Full text of "RichardSwinburne"

See other formats

OXFORD

Providence and the Problem of Evil

This page intentionally left blank

Providence and the Problem of Evil

Richard Swinburne

CLARENDON PRESS • OXFORD
1998

OXFORD

Great Clarendon Street, Oxford 0X2 6DP
Oxford University Press is a department of the University of Oxford
It furthers the University's objective of excellence in research, scholarship
and education by publishing worldwide in
Oxford New York

Auckland Bangkok Buenos Aires Cape Town Chennai
Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata
Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi
Sao Paulo Shanghai Taipei Tokyo Toronto
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States by
Oxford University Press Inc., New York
© Richard Swinburne 1998
The moral rights of the authors have been asserted
Database right Oxford University Press (maker)

First published 1998

All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographcs rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose this same condition on any acquirer
British Library Cataloguing in Publication Data
Data available

Library of Congress Cataloging in Publication Data
Swinburne, Richard.

Providence and the problem of evil / Richard Swinburne.

Includes bibliographical references and index.

1 . Providence and government of God 2. Good and evil.

3. Theodicy. L Title.

BT135.S95 1998 231'.8 — dc21 98-6781
ISBN 0-19-823799-5
ISBN 0-19-823798-7 (Pbk)

Acknowledgements

I have discussed the topic of this book in many lectures and classes in many universities over many years, and I am
very grateful to all those whose comments have forced me to abandon certain lines of argument and encouraged me to
develop others. I have given several graduate classes at Oxford in which I have used earlier versions of all this material;
and most recently I have had the benefit of reactions to it of a graduate class at St Louis University. I am especially
grateful to Eleonore Stump, to Sarah Coakley, and to two referees for the Oxford University Press, both for detailed
criticisms and for very general reactions to some of my moral assumptions. I owe very considerable thanks also to
various secretaries who have typed and retyped versions of this book over the years, and among them especially to Mrs
Fiona Snyckers and Mrs Rosemary Clayton.

Some of my material has been published in various earlier forms. Two and a half chapters of The Existence of God
(Clarendon Press, 1979; rev. edn. 1991) were concerned with theodicy, and I am grateful to the Oxford University
Press for permission to reuse some of that (in a much changed form). I am grateful to the editors and publishers of
four more recent papers for permission to reprint material which has some recognizable connection with chapters of
this book: ‘Knowledge from Experience and the Problem of Evil’, in W. J. Abraham and S. Holtzer (eds.), The
Rationality of Religious Belief (Clarendon Press, 1987); ‘Does Theism Need a Theodicy?’, Canadian journal of Philosophy, 18
(1988), 287-312; ‘Theodicy, our Well-Being, and God's Rights’, International journal for the Philosophy of Religion (Kluwer
Academic Publishers), 38 (1995), 77-93; and ‘Some Major Strands of Theodicy’, in D. Howard-Snyder (ed.), The
Evidential Argument from Evil (Indiana University Press, 1996).

This page intentionally left blank

Contents

Introduction 1

Part I. THE PROBLEM OF EVIL

1. The Need for Theodicy 9

2. Theodicy in Christian Tradition 36

Part II. THE GOOD GOALS OF CREATION

3. Beauty 55

4. Thought and Feeling 60

5. Action 88

6. Worship 117

Part III. THE NECESSARY EVILS

7. The Fact of Moral Evil; and Free Will 131

8. The Range of Moral Evil; and Responsibility 144

9. Natural Evil and the Scope for Response 166

10. Natural Evil and the Possibility of Knowledge 182

11. The Evils of Sin and Agnosticism 199

Conclusion to Part III 223

Part IV. COMPLETING THE THEODICY

12. God's Right 229

13. Weighing Good against Bad 243

Additional Notes 259

Index 265

This page intentionally left blank

Introduction

The Universe with its billions of galaxies, each with billions of stars, many of them probably orbited by the smaller and
colder bodies which we call ‘planets’, has been expanding rapidly for the last fifteen billion years. The ‘Big Bang’, which
began this period of the Universe's history, may have been the beginning of the Universe itself, or the Universe may
have had an earlier history — possibly one without beginning. For the last three billion or so years there have been
animals on the planet Earth, and more recendy humans. Maybe there are animate beings on other planets too, but of
them we have as yet no knowledge. The traditional theist, who believes that there is a God who is all-powerful and
perfecdy good, believes that God created and sustains all this, for supremely good purposes. Some of these good
purposes, he believes, have already been realized or are now being realized. But most traditional theists, and among
them Christian theists, believe that other good purposes are yet to be realized in this Universe or in another one; and
that what is happening in the Universe now is a necessary step towards the realization of these other purposes. These
good purposes include the perfecting of this Universe in all its aspects and the worship of God in the life of Heaven by
those humans who have freely chosen that sort of life. The world is thus, according to Christian theism, the object of
God's providential care — he foresees and meets the needs of his creatures. Some of the good purposes of God
concern all creatures or all humans, and these are the concern of what is called his general providence; others concern
particular individuals — he has a certain good purpose for me and another one for you — and these are the concern of
what is called his special providence.

Such is the Christian doctrine (and that of many other forms of theism). But if God has these supremely good
purposes for the future, why the delay? Why cannot we have their benefits now? And notoriously the world contains
much suffering and other evil which it would seem that God would (in virtue of his perfect goodness) have sought to
prevent, and (in virtue of being all-powerful)

INTRODUCTION

would have been able to prevent. The theist maintains that God (who can do anything logically possible, that is
anything the description of which does not involve a contradiction) could not achieve some of his good purposes
except by means of a delay before they are achieved, and these and other good purposes except by means of allowing
evil to occur. In this book I shall expound and justify this view, especially in the form which it has taken in Christian
theology.

I begin by arguing in Part I that, at any rate in the West in our modern world, most theists need a theodicy, an account
of reasons why God might allow evil to occur. Without a theodicy, evil counts against the existence of God. I shall then
consider in Part II what are some of the purposes which an all-good God might have for the Universe; and point out
that Christian doctrines (of varying degrees of centrality to the Christian tradition) teach that many of them (and
especially the most important ones) are indeed God's purposes. I shall point out that so many of these good purposes
have been, or — if the Christian revelation is true — will be realized in the world. I shall go on, however, to list the
various evils which the Universe contains. I shall then argue in Part III that some of the good states cannot be achieved
without delay and suffering, and that the evil of this world is indeed necessary for the achievement of these good
purposes. I go on to argue in Part IV that God has the right to allow such evils to occur, so long as the goods are
facilitated and the evils are limited and compensated in the way that various other Christian doctrines (of human free
will, life after death, the end of the world, etc.) affirm. And I shall conclude by claiming that the good states which
(according to Christian doctrine) God seeks are so good that they outweigh the accompanying evils.

Theodicy, the defensive element of this book, is territory on which I have ventured before — two and a half chapters of
The Existence of God were concerned with it. But, while continuing to endorse the general approach of that book, I have
come to believe subsequently that theodicy is a considerably more difficult enterprise than I represented it there. This is
because there are so many diverse goods and evils, and they have intricate logical connections with each other: you
cannot have certain goods without certain evils, unless you bring about other goods as well which would bring

Rev. edn. (Clarendon Press, 1991),

INTRODUCTION

about certain evils, which could be avoided but only at the cost of other evils, etc. I assumed in that book that theodicy
does not need to bring in doctrines peculiar to different religions (such as reincarnation in Eastern religions; or life
after death in a new world etc. in Christianity), in order to show that the occurrence of evil does not count against the
existence of God. I am not fully convinced about that any more. In any case most other contemporary humans are a
lot more likely to be convinced if theodicy does bring in such doctrines. So this book invokes relevant Christian
doctrines, such as those mentioned above, which make claims about what God has done and will do.

I seek to show how the whole Christian doctrinal package taken together — the claim that there is a God together with
the claim that he has done and will do certain things for his creatures — faces the difficulty of delay and suffering on the
way to the good things God plans for us. I shall not be producing positive arguments for the truth of these other
Christian doctrines; although I shall occasionally point out with respect to some of them that they are compatible with
our secular knowledge of how the world works and of what is morally good or bad. Positive argument must come
largely from revelation. In so far as there is evidence that God through Christ founded a church as the vehicle of
revelation, and that church has taught these other doctrines, that is the positive evidence that these doctrines are true.
Some of these doctrines are very central to the Christian tradition, such that if you believe that there are any doctrines
revealed by God through Christ's church, you will have to believe these ones. Others, I shall note, are more peripheral:
you could believe that God gave the church the authority to teach doctrines, without believing that these doctrines are
included; more consideration of how to determine what are the exact boundaries of revelation would be needed. So
evidence that God through Christ founded a church as the vehicle of revelation is less strong evidence for these
doctrines. There is no space to discuss the evidence for revelation here ; 2 I seek only to explore the consequences of
supposing certain doctrines to be true, for the central doctrine which is the concern of this book — the doctrine of
Providence . 3

2 For discussion of what would be evidence for a Christian revelation, see my Revelation (Clarendon Press, 1992), esp. pts. 2 and 3.

3 Any evidence (received by a revelation or in any other way) of the truth of some specific doctrines of some religion rival to Christianity could also be evidence in favour of a
doctrine of Providence (though not the Christian doctrine) enabling it to rebut an atheistic objection from the fact of evil. One doctrine which is denied by Christianity but
affirmed by Hinduism and Buddhism, and would be of assistance in this connection, is the doctrine of Pre-existence. Hinduism and Buddhism teach that souls become
reincarnate in different bodies in the next life, according to their behaviour in this — good behaviour leads to a higher life, bad behaviour to a lower one. Present-day humans
are supposed to have had a long series of previous lives. Hence their suffering in this life could be explained as a punishment for misdeeds in the last one. This defence is not
now open to orthodox Christianity — though a few early theologians, including one of the greatest, Origen, held it — see Origen, De Principiis 1.8.4.

INTRODUCTION

Of course a particularly Christian doctrine of Providence is itself a central Christian doctrine, and any evidence for the
revelation through Christ is evidence in its favour. But in view of the obvious initial objections to it, I seek to explore
how with the aid of other Christian doctrines (not subject to initial objections of the same kind) the doctrine of
Providence can be defended against those objections, when given its specifically Christian form. Providence and the
Problem of Evil is the last of four books in which I have been examining the philosophical issues involved in certain
Christian doctrines.

If the only goods in the world were thrills of pleasure and the only bad things stabs of pain, then it would be easy for
God to make a good world with nothing bad in it. And if there was any pain at all — if just one human felt the slightest
toothache — that would be conclusive evidence against the existence of God. But in fact, I shall be claiming, the good
of individual humans (and in so far as they are capable thereof, the good of animals) consists (as well as in their having
thrills of pleasure) in their having free will to choose between good and evil, the ability to develop their own characters
and those of their fellows, to show courage and loyalty, to love, to be of use, to contemplate beauty and discover
truth — and if there is a God it consists above all in voluntary service and adoration of him in the company of one's
fellows, for ever and ever. All that, I shall be arguing, cannot be achieved without quite a bit of suffering on the way.

This book will only convince fully those readers who come to accept many of the moral views advocated in Part II
about which actions and states of affairs are good, and the views about just how good are such actions and states of
affairs advocated in the final chapter of Part IV. I hope that many readers, if they do not already

INTRODUCTION

share those views, will come to accept them in the light of my arguments. But I am very well aware that some of them
are views very different from those taken for granted in the moral climate prevalent today in the Western world; and
many readers may react instinctively against them, sometimes with hostility. I understand that reaction — part of me
shares it. But I believe that the views which I have expressed are true, and are implicit in many of the moral
judgements which even today's Westerners make; and I have tried to draw out the implications. I am well aware,
however, that a radical change of moral stance often requires much reflection on what is good and bad about many
situations, real and imaginary, more than there is space to set out here; and often some harsh experiences as well. But I
hope that readers who do not yet come to share those moral views will come to recognize that a satisfactory theodicy
can be constructed, if those moral views are accepted — and that the grounds for accepting those moral views are very
largely independent of theology.

I can easily understand many deeply sensitive people reacting with horror at the very attempt to show that a loving
God could allow humans and animals to suffer, let alone suffer some of the horrible things that happen on Earth. But
this book constitutes a plea to such people, whose sensitivities I deeply respect, to ask themselves just what in detail
love might amount to in an all-powerful being who brings all creatures into being out of nothing, concerned to be
generous — not just to already existing humans for the short space of an earthly life, but to many possible kinds of
creatures for unending time — and generous with the best and deepest of gifts. I suggest that long reflection on this will
make it less and less obvious that some significant suffering for the very short period of an earthly life is ruled out.

This page intentionally left blank

I The Problem of Evil

This page intentionally left blank

1 The Need for Theodicy

My soul is weary of my life; I will give free course to my complaint; I will speak in the bitterness of my soul. I
will say unto God, Do not condemn me; Show me wherefore thou contendest with me. Is it good unto thee that
thou shouldest oppress, that thou shouldest despise the work of thine hands, And shine upon the counsel of the
wicked?

(Job 10: 1-3)

The Problem of Evil

I understand by ‘God’ a being who is essentially eternal, omnipotent, omniscient, creator and sustainer of the Universe,
and perfectly good. 4 An omnipotent being is one who can do anything logically possible, anything, that is, the
description of which does not involve a contradiction: such a being could not make me exist and not exist at the same
instant, but he could eliminate the stars or cover the Earth with water just like that. An omniscient being is one who
knows everything logically possible for him to know, anything the description of his knowing which does not involve a
contradiction. He would know everything that has happened, everything that is happening or could happen. But, in my
view (to be explored more fully in Chapter 7) he will not necessarily know everything that will happen unless it is
already predetermined that it will happen. For there is a logical inconsistency in supposing that any being knows
necessarily what is yet to happen when that has yet to be determined (i.e. when it is not already fixed by its causes). But
if the omniscient being is God and so also omnipotent, it will be through

4 See Additional Note 1 .

THE PROBLEM OF EVIL

his own choice that there is anything not already predetermined and so that there is any limit to his knowledge.

I shall henceforward normally contrast the ‘good’ with the ‘bad’ rather than with the ‘evil’, both when talking about the
actions of agents and their characters, and also when talking more generally about states of affairs — though an agent of
very bad actions and character is appropriately called evil. ‘States of affairs’ is a term which I use in the widest sense to
include the things that happen to people as well as the intentional actions they perform. Pains and other suffering are
bad states of affairs, but it is odd to call them evil (as I did in the Introduction), even if some agent causing them or
allowing them to occur would be doing an evil act. Although the problem with which we are concerned is called ‘the
problem of evil’ (and so those words form part of the title of this book) it is really the problem of the existence of bad
states of affairs, such that (it is claimed) it would be bad for an agent who could prevent them to allow them to occur.
But though I shall contrast good states with bad states as such, in deference to the terminology in which our concerns
are discussed, I shall, when I come to discuss bad states in detail in Part III and need to distinguish them into two
groups, subdivide them in the traditional way into ‘moral evils’ and ‘natural evils’ (using the term ‘evil’ only when
preceded implicitly or explicitly by the relevant adjective).

I understand ‘moral evil’ as including all bad states caused deliberately by humans doing what they believe to be bad,
and especially wrong (or allowed to occur by humans negligently failing to do what they believe to be good, and
especially what they believe to be obligatory) and also the bad states constituted by such deliberate actions or negligent
failure. It includes the pain I deliberately inflict on you, the pain of the disease which I negligently allow you to contract
by allowing you to come into contact with an animal of mine who suffers from it, and the bad state of my deliberate
infliction or negligent omission. There is also moral evil constituted by humans doing what they believe to be bad
where no suffering results, e.g. telling a lie, breaking a promise, or attempting to inflict suffering without success.
(Doing something bad which the agent does not believe to be bad counts, on my definition, as natural evil.) I
understand by ‘natural evil’ all bad states not thus deliberately produced by human beings, or through negligence
allowed by human beings to occur. Natural evil thus includes all the trail of

THE NEED FOR THEODICY

suffering which disease and accidents unpreventable by man bring in their train. It also includes the bad desires with
which we find ourselves: the temptations to take more than our fair share of the world's goods, to lie and deceive for
the sake of our reputation. Among natural and moral evils are the sufferings of non-human sentient beings, of which
the only ones known to us are the higher animals.

While the adjective ‘moral’ in front of the word ‘evil’ thus distinguishes a certain kind of bad state from others, there is
a wider sense in which all the ‘good’ and ‘bad’ states with which we are concerned in this book are ones which are
morally good or bad. In this wider sense (which will always be the one used, except where the adjective ‘moral’
precedes the word ‘evil’) the ‘moral’ is the ‘overall’ or ‘overriding’. A morally good act is one which is overall a good
act. It might be bad from the point of view of aesthetics or etiquette that I should support a project for building cheap
housing or speak to someone to whom I had not been introduced, but if it is overall a good act, then it is morally good.
The morally best act for me to do is the one which is overall the best act, the one whose goodness overrides that of the
goodness of other acts. Among (morally) good acts are obligations — to keep promises or not to wound someone; but
there are good acts which go beyond obligation — to give money needed to feed me at a proper level, in order to feed
someone else at a proper level, for example. A morally bad act is one which is overall bad; and among such acts are
ones which are wrong, i.e. obligatory not to do. A wrong act which wrongs God is a sin. I shall use a ‘right action’ in
the same sense as a ‘good action’. (I contrast a ‘right action’ with someone having ‘a right’ to do some action, which is
to be understood as saying that it is morally justified or permissible, i.e. not wrong, to do the action.) I shall mean by
saying of some state of affairs that it is morally good, that it is good (on balance, overall) that it occur; and by saying
that it is morally bad, that it is bad overall that it occur. Natural evils are thus morally bad states. I apologize for this
double usage of ‘moral’: it is alas current in the philosophical literature, and it will, I think, be less confusing to the
reader to stick with it (having made clear when ‘moral’ is being used in one sense rather than another) than to
introduce totally new terminology.

When I write (as in the last paragraph) of ‘good’, ‘bad’, etc. acts, my concern — unless I specify otherwise — is with
objectively good,

THE PROBLEM OF EVIL

bad, etc. acts. An act is objectively good (bad) if it is good (bad) in its nature or consequences, apart from what the
agent believes about it. But an act may have a nature or consequence of which the agent is ignorant. An act is
subjectively good (bad) if its agent believes that it is good (bad) in nature or consequences. The subjective goodness of
an agent is a matter of the subjective character of his actions. A subjectively perfectly good agent will do many
subjectively good acts and no subjectively bad acts. Such an agent notoriously may do many acts with many bad
consequences, unforeseen by him but foreseeable by more knowledgeable agents. However, if the agent is also
omniscient, he will know all the consequences of his actions which it is logically possible for him to know. But unless it
is already predetermined what consequences some action will have, he will still not (in my view — to be explored more
fully in Chapter 7) know incorrigibly what consequences that action will have. He will, however, know what
consequences that action is naturally likely to have, the extent of the (indeterministic) bias in nature towards certain
consequences rather than others. 5 1 shall in future understand the objective goodness of an action more precisely than
on my previous definition: as the goodness it has in virtue of the objective features of the world at the time it is
performed. Its goodness in respect of its consequences is a matter only of its objectively likely consequences, not of its
actual consequences. And not every act which has some predetermined bad consequences is objectively a bad act; its
other consequences or aspects of its nature may make it overall a good act. An objectively perfectly good and
omniscient agent will do many objectively good acts and no bad acts. (Unless stated otherwise, my concern hence-
forward is with the objective goodness of actions and agents.)

I believe that Western religion (i.e. Christianity, Judaism, and Islam) has understood by God roughly a being of the kind
which I have described; and it is that understanding of ‘God’ which gives rise to the ‘problem of evil’, initially in the
form of an apparently

To the extent to which there is a bias in nature, we say that the earlier state or action makes a later state naturally or physically probable or likely. Probability of this kind is to
be contrasted with evidential probability, which is a measure of the extent to which certain evidence makes it likely that some event occurred; and with statistical probability,
which is a measure of the proportion of certain kinds of events in wider classes (e.g. of tosses of heads among tosses of a coin). On all this, see e.g. my An Introduction to
Confirmation Theory (Methuen, 1973), chs. 1 and 2 .

THE NEED FOR THEODICY

conclusive argument against the existence of God. For, the atheist's argument goes:

1 . If there is a God, he is omnipotent and perfectly good.

2. A perfecdy good being will never allow any morally bad state to occur if he can prevent it.

3. An omnipotent being can prevent the occurrence of all morally bad states.

4. There is at least one morally bad state.

So (conclusion): There is no God.

The argument purports to be — and fairly obviously is — a valid deductive argument, i.e. one in which the conclusion
simply draws out what is implicit in the premisses; one in which if you admit the premisses but deny the conclusion,
you contradict yourself. This is to be contrasted with a correct inductive argument, which I shall understand for
present purposes as one in which the premisses make the conclusion probable. A sound deductive argument is one
which is valid and in which all the premisses are true. I understand by a sound inductive argument one which is correct
and in which all the premisses are true.

The first premiss of the above argument simply states some of what is involved in the Western understanding of God,
as I have spelled it out. If God's goodness were supposed to be other than moral goodness, then it might be no
objection to his existence that there is pain and other suffering. But, despite the fact that some philosophical
theologians 6 have attempted to expound God's goodness in non-moral ways, it seems to me deeply central to the
whole tradition of the Christian (and other Western) religion that God is loving towards his creation and that involves
his behaving in morally good ways towards it. There is no doubt more to loving someone than not kicking them in the
teeth. But it does (barring special considerations) seem to involve at least not kicking them in the teeth. Western
religion has always held that there is a deep problem about why there is pain and other suffering — which there would
not be if God were not supposed to be morally good. Again, God is supposed to be in some way personal, and a
personal being who was not morally good would not be the great being God is

Thus Brian Davies claims that ‘Theologians have taught that God is good without holding that his goodness is that of a morally good agent’ {An Introduction to the Philosophy of
Religion, 2nd edn. (Oxford University Press, 1993), 48) .

THE PROBLEM OF EVIL

supposed to be. Hence, in view of the fact that my concern in this chapter is with the existence (or not) of the sort of
God with which Western religion is concerned, this understanding of the definition of God must stand.

Premiss 4 is obviously true. Theists have sometimes challenged premiss 3 on the ground that to be omnipotent a being
has only to be able to do the logically possible, and to prevent the occurrence of all morally bad states of affairs is not
logically possible; maybe he could prevent the occurrence of any particular bad state, but not of them all taken
together. This challenge does not look very plausible: it does look as if it is logically possible to make a beautiful world
in which the only conscious beings are predetermined to live lives of unalloyed pleasure; and then surely there would
be no bad states. It might be that in such a situation, without the existence of contrasting bad states, the world's
inhabitants would not realize that their world contained nothing bad — but it would be true all the same. Anyway I see
no reason to suppose that the inhabitants could not realize that their world contained nothing bad; they could have the
concept of bad (they could be born with an ability to recognize bad states, if they were to occur) without there actually
being any instances of bad states. Again, if the world's badness were supposed to be other than moral badness, it might
not be logically possible for God to prevent it. That might be so, for example, if the bad were understood in one old-
established way of dealing with the problem of evil as privatio boni, taken simply as the absence of some good. For then
it would not be logically possible for God to prevent the occurrence of all bad, since whatever world God makes, he
can always make a better one. 7 However many persons God made, for example (even if it were an infinite number), he
could always add another one (at sufficient distance from others so as not to crowd them out) and then there would be
a better world. It follows that the objection from the existence of bad states of affairs to the existence of God is not
that he did not create the ‘best of all possible worlds’, for to do that is not logically possible: there is no best of all
possible worlds. 8 The objection to God's existence is

7 The rejection of the third premiss on these grounds is in effect the solution to the problem of evil given by G. Schlesinger, Religion and Scientific Method (D. Reidel, 1977), pt. 1

8 Leibniz classically claimed that God would make ‘the best of all possible worlds’, and indeed had done so. ‘If there were not the best among all possible worlds, God would
not have produced any 5 (G. W. Leibniz, Theodicy, trans. E. M. Huggard (Routledge & Kegan Paul, 1951), §8) . But, alas, in Leibniz's sense of ‘possible world’ God cannot fail
to make actual some possible world. For not to create any creatures is to make a world empty of all things except God. And that too, on my argument, would be to make a
world less good than some other world. Aquinas, by contrast, rightly claimed: ‘God could make other things, or add other things to those he has made, and there would be
another and better universe’ ( Summa Theologiae , la. 25. 6 ad 3).

THE NEED FOR THEODICY

rather that there are some actual morally bad eliminable states; moral badness is not just the absence of moral
goodness. So premiss 3 must stand.

The normal challenge to the atheist's argument and surely by far the most plausible challenge is to premiss 2, and I
shall be concerned with that alone in the main argument of this book. I suggested earlier that a perfectly good being
will do many (morally) good actions and no bad ones. The reader might have thought that this was too modest an
account to constitute a definition of perfect goodness. Surely a perfectly good being, it might be thought, will, in any
situation, always do the best action; or, at any rate, if there are a number of equal best actions, do one of them. I agree
that the definition should be made more precise by adding to it that a perfectly good being will always do the best or an
equal-best action — where there is one. But the point made above that there is no best of all possible worlds has the
consequence that for an omnipotent being, there is sometimes no best or equal-best action. For an agent of limited
power, a human being for example, will have available to him only a limited range of actions; and one of them may
indeed be the best, or some of them be equal best. It may, for example, be the best action for me to do to give a lot of
money to some charity for the relief of famine, but equally best to give it to the Red Cross, or to give it to Oxfam or
Christian Aid. We could understand the perfect goodness of such a being as its doing always the best or an equal-best
action. But we cannot understand perfect goodness generally in this way; and we cannot understand the perfect
goodness of a God who is omnipotent in this way. For he will so often have available to him an infinite range of
actions, each of which is less good than some other action he could do. Thus for any world of conscious agents which
God could create there is, I suggested, a better one, for example one obtained by adding one more conscious agent.
And so among the actions of creating conscious agents there is no best (or equal best). What goes for creating
conscious agents goes also for creating inanimate things. And no doubt, though I do

THE PROBLEM OF EVIL

not need to argue this here, for much else too, for the kinds of pleasure and opportunities of service he gives to beings.
For such a situation we cannot say anything more positive about what a perfectly good God will do.

Note, however, that although God may not always be able to do a best act, he will always be able to fulfil his
obligations. For obligations arise either from commitments entered into voluntarily (e.g. one's obligations to one's
children) or from benefits received (e.g. one's obligations to parents). The only benefits which God receives are ones
which he chooses to receive. Hence all God's obligations are ones voluntarily entered into. And God can ensure that
they form a limited set such that he can fulfil all of them; and since failing to fulfil one's obligations would be a morally
bad act (of omission), God will always fulfil his obligations. 9

Now what is wrong with premiss 2 is that it is not always a bad act to bring about or to allow to occur a bad state of
affairs. For it may be that the only way in which an agent can bring about some good state is by bringing about first (or
simultaneously) some bad state, or by allowing such a state to occur. Thus the only way in which a human parent can
get his child's teeth repaired may be by taking him to the dentist and allowing pain to be inflicted upon him; and
sometimes the only way in which he can reform the child may be to punish him by deliberately inflicting suffering
upon him. But the human parent is none the less good for taking the child to the dentist, or for punishing him. So
premiss 2 is false.

The atheist is able, however, to put forward a more sophisticated version of premiss 2. He may claim, plausibly, that it
is always a bad act to bring about a bad state of affairs, or, more generally, allow one to occur, where the agent can
prevent it, unless thereby the agent makes possible a good state of affairs which would not otherwise be possible. And
even so, it is only under restricted conditions that a perfectly good being will allow a bad state of affairs for the sake of
the good that it makes possible. We need to qualify premiss 2 in various further ways. First, for some bad states, even if
that bad state is a necessary condition of some good, agents would have no

Obligations are, I suggest, always obligations to someone else. They are debts, what one owes; and the notion of a debt to oneself makes no sense. So one cannot wrong
oneself, only someone else. One may of course harm oneself, do something bad to oneself — but it only constitutes a wrong, if it is a wrong to someone else — e.g. to my
parents, who have lavished care on me, or to God. On this, see my Responsibility and Atonement (Clarendon Press, 1989), various pages up to p. 77 .

THE NEED FOR THEODICY

right ever to bring it about or even to allow it to occur for the sake of some good. (That is, it would never be morally
permissible for them to bring it about or allow it; they would never be morally justified in so doing.) A beating may do
some child much good, and that amount of suffering might even be logically necessary to effect that particular good;
but, unless I am the child's parent or teacher, I do not have the right to beat the child for the sake of that good. And
there are other bad states which I do not have the right to allow to occur (if I can easily prevent them), e.g. some
stranger strangling my baby in order to use its organs to save the life of some other baby. Since one is obliged not to do
that which one does not have the right to do and God always fulfils his obligations, the bad states which he allows to
occur must be ones which he has the right to allow to occur. Take, for example, the traditional free-will defence which
we will consider in detail later in the book. This claims that many of the bad states which God allows to occur are ones
which humans freely choose to inflict on each other, that it is a good thing that humans have such freedom, and the
bad states — e.g. the pains and other sufferings which humans inflict on each other — are the price which is paid for that
freedom. If this defence is to work, God must have the right to allow the human victims to suffer in the circumstances
in which they do suffer.

Secondly, the atheist must acknowledge that the logically necessary condition of the good may not be the actual
occurrence of the bad, but some agent P having the power to cause the bad state to occur without being caused to
exercise that power or being caused to refrain from exercising that power. Consider again the free-will defence. This
claims that humans have the power, which I shall call libertarian free will, of choosing whether or not to bring about
effects without being subject to causes which determine how they will exercise that power; and that among the effects
which they can choose whether or not to bring about are happiness or suffering for others. This good of their having
this choice and their choice making a difference (free will plus efficacy, which I shall call efficacious free will) has as its
necessary condition that their actions are not predetermined and that if they choose to bring about suffering, suffering
will result. But the actual occurrence of the suffering is not a necessary condition of the good. (Humans might simply
always choose the good — but they would not have efficacious free will if God ensured that only their good choices
were efficacious.)

THE PROBLEM OF EVIL

Note that the necessary condition of the good might not be allowing the particular bad state in question but allowing
some state of affairs just as bad or worse. It is no objection to theism that God allowed something bad to occur, in
order to achieve some good, when he could have achieved the good by allowing a different but equally bad state to
occur instead. Thus allowing you to suffer is not a necessary condition of my having efficacious free will to choose
whether or not to cause suffering, but allowing someone to suffer is such a condition.

Thirdly, it must be the case that the perfecdy good agent does all else that he can (compatible with allowing no further
bad state to occur) to bring about the good. If allowing some bad to occur makes possible some good, but God does
not do all else that he can to bring about the good, then he should not have allowed the bad state to occur. Thus if
God would be justified in allowing you to suffer in order that I may exercise efficacious free will in choosing to
respond to it with sympathy, but God does not in fact give me such free will, then clearly the defence does not work.
Or if God would be justified in allowing a certain amount of suffering on Earth, if it is compensated by happiness in
the afterlife, but God does not give us an after-life, then again the justification does not work.

And finally, we need some sort of comparative condition. It cannot be as strong as the condition that the good aimed
at, G, be more of a good state than the bad means, E, is bad. For obviously we are often justified, in order to ensure
the occurrence of a substantial good, in risking the occurrence of a worse state. To save one life, we are often justified
in embarking on a course of action where there is a small risk of two lives being lost but quite a good chance of no
lives being lost. But it must be that if the perfecdy good agent intentionally brings about E for the sake of G, G must be
at least a bit more of a good state than E is a bad state. And if he allows E to occur but does not bring it about (e.g.
brings it about that chance or some other free agent determines whether E occurs), then the more probable it is in
these circumstances that E will occur, the closer E must be to being a less bad state than G is good. A formal way of
capturing this condition, if we measure good consequences by positive units and bad consequences by negative units, is
that the action must overall have a positive expected value . 10

The expected value of some action is the sum of the values (positive or negative) of its possible consequences, each multiplied by the probability of its occurrence if the
action is done. Anyone who thinks that there are no limits to God's knowledge of the (to us) future actions of free agents (or consequences of chance processes), can clearly
operate with the simpler formula, that the positive value of G must exceed the negative value of E . (Some may reasonably hold that all that is required is a non-negative, i.
e. positive or zero, expected value. But I am putting the requirement in the simpler form, since the difference between zero and a very tiny positive balance is so small that
virtually nothing will turn on this.)

THE NEED FOR THEODICY

So given these amplifications, the atheist should agree to abandon premiss 2, A perfectly good being will never allow
any morally bad state to occur if he can prevent it’, and replace it by the more careful premiss 2*, A perfectly good
being will never allow any morally bad state E to occur if he can prevent it, unless (i) allowing E to occur is something
which he has the right to do, (ii) allowing E (or a state of affairs as bad or worse) to occur is the only morally
permissible way in which he can make possible the occurrence of a good state of affairs G, (iii) he does all else that he
can to bring about G, and (iv) the expected value of allowing E, given (iii), is positive.’ I shall in future sometimes
summarize premiss 2* as A perfectly good being will never allow any morally bad state to occur if he can prevent
it — except for the sake of a greater good’, but understand by this the careful spelling out which I have just provided.

The atheist will now claim that the earlier argument can be reconstructed, using premiss 2* instead of 2, as a valid
deductive argument from the existence of many a particular evil E to the non-existence of God, as follows:

1 . If there is a God, he is omnipotent and perfectly good.

2. A perfectly good being will never allow any morally bad state E to occur if he can prevent it, unless (i) allowing
E to occur is something which he has the right to do, (ii) allowing E (or a state of affairs as bad or worse) to
occur is the only morally permissible way in which he can make possible the occurrence of a good state of
affairs G, (iii) he does all else that he can to bring about G, and (iv) the expected value of allowing E, given (iii),
is positive.

3. An omnipotent being can prevent the occurrence of all morally bad states.

4. There is at least one morally bad state e which is such that either God does not have the right to allow e to
occur, or there is no good state y, such that allowing e (or a state at

THE PROBLEM OF EVIL

least equally bad) to occur is the only morally permissible way in which God can make possible the occurrence
of y, that God does all else that he can to bring about y, and that, given the latter, the expected value of
allowing e is positive.

So: There is no God.

While accepting the new second premiss, the theist now denies the new fourth premiss. He claims that there are no
morally bad states of the kind specified there.

Since all things are possible for an omnipotent being, except the logically impossible, the theist's defence is then that,
compatibly with his perfect goodness, God may allow a bad state E to occur, caused either by himself or some other
agent, if (and only if):

(a) God has the right to allow E to occur.

(b) Allowing E (or a state as bad or worse) to occur is the only morally permissible way in which God can bring
about a logically necessary condition of a good G.

(d) The expected value of allowing E, given (c), is positive.

The theist claims that all these criteria are satisfied with respect to each bad state which there is in the world. The
atheist denies this, and certainly it is not obvious to an impartial inquirer that the new premiss 4* is false. Some of this
world's horrors seem at first sight to be such that God does not have the right ever (or in existing circumstances) to
allow them to occur, or that there is no comparable good which they make possible.

Does the Theist Need to Prove His Case?

Having thus clarified on what the dispute turns, I now seek to investigate whether the theist is entitled to hold that all
the morally bad states of affairs are such that (by our four criteria) God could, compatibly with his perfect goodness,
have allowed them to occur, without having any view about how those criteria are satisfied. Can he just say: God must
have some good reason for allowing bad states to occur, though we cannot see what they are? Or does the theist, in
order justifiably to maintain his belief, need to be able to

THE NEED FOR THEODICY

show (to his own satisfaction) how, if there is a God, our four criteria (a, b, c, and d) are probably satisfied in respect of
all the world's morally bad states (namely, that probably they are such that a perfectly good God could allow them to
occur)? I shall understand by an attempt to show the latter, the construction of a theodicy (theou-dike, ‘a justification of
God’) for that state. I thus understand by a ‘theodicy’ not an account of God's actual reasons for allowing a bad state
to occur, but an account of his possible reasons (i.e. reasons which God has for allowing the bad state to occur,
whether or not those are the ones which motivate him ). 11 The world's morally bad states fall into kinds (e.g. pain
deliberately caused by humans, pain allowed to occur through human negligence, pain caused by natural processes
uncontrollable by humans, malevolent desires, grief, etc.). If the theist can provide for states of each kind a reason why
God could justifiably allow a state of that kind to occur — e.g. pain deliberately caused by humans being justified in
terms of the good of humans having a free choice of whether or not to cause pain deliberately to others — he will have
provided an adequate total theodicy. But does the theist need to have available a theodicy which shows to his
satisfaction (and so in the face of those objections from the atheist of which he is aware) how the four criteria are
satisfied in respect of each kind of morally bad state?

Talk of the theist being ‘entitled’ or ‘rational’ to hold, or ‘justifiably’ (i.e. epistemically, not morally) maintain, this or that
belief can be understood in many different ways. There are internalist senses in which the justification of a belief is a
matter of its right relation to other internal states of the subject (or something in the nature of the belief itself); and
‘externalist’ senses in which justification of a belief is a matter of its being produced in the right way,

I believe this to be the traditional use of the term ‘theodicy I * * * 5 . Alvin Plantinga has confined the use of ‘theodicy 5 to a doctrine about what are God's actual reasons for allowing

evil to occur. (See his The Nature of Necessity (Clarendon Press, 1974), 192 .) This use has been adopted by many other writers. What I call a ‘theodicy 5 is closer to what

Plantinga calls a ‘defence 5 . The difference is that a Plantinga-type ‘defence 5 is required merely to show that the existence of the bad state is compatible with the goodness and

power of God — in effect that there is a G for which clauses (a ), (b ), and (d ) are satisfied. I require also that it be shown that (c ) is probably satisfied for such a G . This is

vital if the atheist's argument is to be resisted. In those cases where the claim of (c ) is a central claim of Christian revelation (e.g. that there is life after death), I am assuming
in this book that arguments for the probable truth of this revelation (e.g. those adumbrated in my Revelation (Clarendon Press, 1992) ) are successful.

THE PROBLEM OF EVIL

e.g. by a process which normally produces true beliefs. My concern here is solely with that (subjective) internalist sense
in which a belief is justified (the subject is ‘entitled’ to hold it, is rational in holding it) if and only if either (by the
subject's own standards of probability) it is rendered probable by the subject's other justified beliefs 12 or it is properly
basic. A belief is properly basic if it is probably true without deriving this probability from any other belief. For an
internalist this latter probability will not come from anything external to the subject; rather, a properly basic belief is
one which is probably true because of its content alone, e.g. because it is a belief about what the subject is now
perceiving . 13 (I shall have more to say about this notion shordy.) Loosely, people are justified

This first clause would need to be stated rather more precisely in a work of epistemology. My precise statement would involve claiming that for a subject to believe a
proposition p is to believe p more probable than one or more incompatible propositions — normally (but not always) the negation, not -p . But since it is not always clear,
when someone is said to believe some proposition, what is the contrasting proposition, it is not always clear what is meant by the claim that he or she believes To believe p
is normally to believe p more probable than not -p , but not always. To believe an explicit conjunction (# and r ), for example, may be only to believe (# and r ) to be more
probable than (# and not-r ), and more probable than (not-# and r ), and more probable than (not-# and not-r ). For the belief to be justified is then for the believed
relations actually (by the subject’s own standards of probability) to hold. Since the normal contrasting proposition is the negation, not-/) , it then follows that normally for a
belief to be justified is for it to be more probable than not, i.e. probable. Where the probability is derived from the subject's other beliefs rather than from its properly basic
status, we then get the statement in the text The sense of ‘rational’, ‘justified’, etc. with which I am concerned here is the one which I called ‘rationality ’ in my Faith and

The notion of ‘properly basic’ beliefs came into philosophical discussion through the work of Alvin Plantinga (see his use of it in his essay ‘Reason and Belief in God’, in A.
Plantinga and N. Wolterstorff (eds.), Faith and Nationality (University of Notre Dame Press, 1983) ). For Plantinga a ‘properly basic belief’ is merely one which the subject is
justified in holding quite apart from any support which it might gain from other beliefs. (Plantinga usually talks of the subject being ‘rational’ in holding the belief; or the
belief being ‘acceptable’. ‘Justified’ is my word.) On this definition it might still be required (by an internalist theory) that for some belief to be properly basic, the subject
should have certain other appropriate internal states, such as sensations (e.g. for him to have a properly basic belief that he is seeing a desk, it might be necessary that there
be a certain pattern of colour in his visual field). However, it seems to me that for an internalist, whether a subject is justified in holding a belief (and so not at fault
epistemically) can depend on what sensations he has only in so far as he is aware of them, i.e. has beliefs about them. And so for simplicity's sake (in a work not primarily
concerned with epistemology), I shall assume here and subsequently that, for an internalist, whether a basic belief is properly basic depends only on its content and not on
any other internal states. And then the issue is whether all basic beliefs are properly basic, or only ones with certain sorts of content.

THE NEED FOR THEODICY

in holding their beliefs to the extent to which they fit with their other beliefs or are forced upon them by their beliefs
about what they are experiencing. People whose beliefs do not fit together are epistemically at fault, because sorting out
our beliefs so as to make them fit together better is something we can do something about. It is desirable that we
should sort out our beliefs in this way because they are more likely to be true if they are rendered probable by our
evidence (i.e. our properly basic beliefs); and our standards of what makes what probable are the ones which we believe
to be the correct ones. But we have access to no reason for conforming our beliefs to some external standards of
justification unless it follows from beliefs which we have or acquire that those standards are correct . 14 We cannot be
criticized for failing to conform our beliefs to standards and facts to which we have no access; and our concern here
with a subject's entitlement to a belief is a concern with whether he is at fault in holding it.

It follows that whether a theist needs a theodicy for every known bad state in order justifiably to believe that there is a
God, despite the ‘argument from evil’, depends on what else he believes and how strongly he believes it. There may
indeed be theists who need no theodicy, but I shall be claiming that most of us do need a theodicy, and for the rest of
the book I shall be endeavouring to provide one.

An atheist's claim with respect to some bad state that the criteria are not satisfied may turn on a purely moral issue, or
it may turn on an issue of non-moral contingent fact. Thus if he claims that a child's suffering can never satisfy
criterion (a), since God does not have the right to allow a child to suffer for the sake of some good however great, the
issue turns on a purely moral issue. But if he claims that the Tree-will defence’ does not work because, although God
would be justified in allowing one human to cause another to suffer if humans had libertarian free will, as a matter of
fact they do

Theories of a subject being justified in holding some belief B often add a condition to the effect that whatever grounds provide the justification for B must be causally
operative in sustaining that belief. The belief must be ‘based’ on those grounds. But whether the justifying grounds (i.e. the other beliefs which make B probable) are
causally operative in sustaining B can only be something of which the subject can take account if he has a belief C about it. If he does, then that belief C can play its role in
raising or lowering the probability of B . Otherwise the ‘basing’ condition must be ignored by a purely internalist theory of justification which is concerned only with a
justification which is accessible to the subject. We shall see the need for a ‘basing’ clause when developing a largely internalist theory of knowledge (as opposed to
justification) in Ch. 4 .

THE PROBLEM OF EVIL

not have libertarian free will, then the issue turns on whether criterion (c) is satisfied in such cases — whether God has
provided what would justify him in allowing the bad state in question, a matter of contingent non-moral fact.

In order to distinguish between the two kinds of challenge, it is important for me to establish at this stage a position on
the status of moral judgements. I hold that they have truth-value: some are true and some are false. I do not need to
argue for that aspect of my position in this context, since anyone who thinks that the occurrence of bad states raises
for theism the ‘problem’ which I have described must think this; or must, in order to show that theism is at odds with
some evident fact, have temporarily assumed it. For there could only arise an issue whether certain bad states were
compatible with the existence of a good God if goodness and badness were properties which belonged to persons,
actions, and states of affairs, and judgements which affirmed or denied their possession had a truth-value. If moral
judgements were mere expressions of emotion, they could not be incompatible with any factual claims.

Now the moral goodness (or otherwise) of particular actions is supervenient on non-moral features of those actions.
That is, judgements such as ‘You ought to fight for France’, ‘It is good to give money to Oxfam’, derive their truth-
value (whether true or false) from contingent states of affairs describable in non-moral terms, i.e. (in Moorean
terminology 15 ) ‘natural’ states. If the first were true, what would make it true would be such things as you being a
French citizen, and France being attacked by an enemy who seeks to rule her; what would make the second true would
be such things as that Oxfam feeds the starving in Africa but does not have enough money for this purpose. But once
all the natural circumstances which make the moral judgements true are set out, that in those circumstances (fully
described) you ought to fight for France, or it is good to give to Oxfam, if the moral judgements are true, they could
not but be true; they would be necessary truths (i.e. ones which held in any logically possible world). There could not
be a world identical with the actual world in all natural respects but in

15 In Principia Ethica (Cambridge University Press, 1903) G. E. Moore distinguished non-natural ‘objects’ and ‘properties’ (the goodness, rightness, ought-to-be-doneness of
things) from natural objects and properties (almost all other properties of things, including whether they are wanted, desired, commanded, or will make people happy).

THE NEED FOR THEODICY

which different moral judgements were true (e.g. a world identical with our world in all natural respects but in which
murder and torture — in our world morally bad actions — were morally good actions). No doubt particular necessary
moral truths, such as ‘In circumstances C you ought to fight for France’, derive their truth from more general moral
truths such as ‘Everyone ought to fight for his country when it is attacked by enemies who wish to rule over it’, but my
argument does not depend on such derivability. My only point is that, if moral judgements are ever true, and those
which are contingently true depend for their truth on natural facts, then the judgements that they hold when those
natural facts hold must be necessary. Hence contingent moral truths derive their truth from contingent natural truths
and necessary moral truths. The basic moral truths are necessary truths.

It might be that among the necessary moral truths are truths to the effect that whatever God commands humans to
do, humans ought to do — within such-and-such limits, or without limits. But even if whatever God commands
becomes a human obligation, that will be so in virtue of a necessary truth connecting the natural fact of God
commanding and the non-natural fact of an obligation to obey.

Now premiss 2* would seem (if, as I have argued, it is true) to be a necessary truth. Its truth does not depend on
whether or not there is a God, or apparently on anything else. (Note too that its truth is unaffected by whether or not
you suppose God's rights to command to be unlimited.) Premiss 3 is also a necessary truth — deriving from the nature
of omnipotence (as defined on p. 3) and moral badness; it holds whether or not there are any omnipotent beings or
bad states of affairs. Premiss 1 is also a necessary truth; it just says what is involved in there being a God, of the sort I
defined. Premiss 4*, affirming the existence of bad states of a certain sort, is not, however, a necessary truth; it is a
contingent matter whether there are such bad states. But if an argument from a contingent truth and general necessary
truths to some conclusion is valid, so is the argument without the necessary truths explicitly stated — necessary truths
being entailed by all other propositions and so already entailed by the other premisses. (I am using ‘p entails q in the
wide sense that q holds in every possible world in which p holds.) Hence whether the atheist has a sound argument
from his premisses to the non-existence of God turns solely on whether premiss 4* is true.

THE PROBLEM OF EVIL

So when an atheist claims that there is some bad state E which does not satisfy the four criteria, his objection may turn
solely on necessary truths of morality. If he claims that criterion (a) could never be satisfied with respect to E (that God
never under any conditions has the right to allow E to occur), or that (b) is not satisfied because there is no good which
E makes possible, he claims that there is an E which a perfectly good God would never bring about. Once E is
described (e.g. a child suffering), then the issue of the truth of premiss 4* turns solely on whether certain proposed
moral principles are necessary moral truths or not. 16 Alternatively, as we have seen, the atheist may admit that God
would be justified in allowing E to occur under certain contingent conditions F, but claim that F do not in fact hold.
The truth of premiss 4* then turns on both necessary and contingent issues.

Either way, the question remains what is the rational response to the suggestion that there is a bad state E with respect
to which premiss 4* is true (i.e. such that either for moral or contingent reasons, criteria (a), (/;), (c), or (d), which would
justify God in allowing E to occur, are not all satisfied).

A supreme principle which covers the justification of belief (in our present internalist sense) is surely the principle
which has been called the Principle of Credulity 17 — that, other things being equal, it is probable and so rational to
believe that things are as they seem to be (and the stronger the inclination, the more rational the belief). By ‘seem’ (or
‘appear’) I mean ‘seem epistemically’; the way things seem epistemically is the way we are initially inclined to believe
that

If the atheist is right in his views about the necessary moral truths, and nothing turns on anything contingent, he will then have a valid deductive argument from true
premisses (including one reporting the occurrence of E ) to the non-existence of God. It seems to be generally agreed by atheists as well as theists that what is called ‘the
logical problem of evil’ has been eliminated, and all that remains is ‘the evidential problem’. See e.g. Paul Draper, who writes that he ‘agrees with most philosophers of
religion that theists face no serious logical problem of evil’ (‘Pain and Pleasure: An Evidential Problem for Theists’, Nous, 23 (1989), 331-50: 349 ). But whether that is so
depends on what we understand by ‘the logical problem’. It has not been shown to the satisfaction of atheists that there is no valid deductive argument from the existence of
certain evident bad states E (via some necessary moral truths) to the non-existence of God. It has been shown merely that there is no such valid deductive argument
evident to theists, who dispute the validity of any such argument by disputing the necessity of the relevant purported necessary moral truths.

For a much fuller and so more adequate discussion of the Principle of Credulity, as it applies to apparent perceptual experiences, see my The Existence of God ' rev. edn.
(Clarendon Press, 1991), ch. 13 .

THE NEED FOR THEODICY

they are . 18 We find ourselves with involuntary inclinations to belief; in the absence of reasons against going along with
such an inclination the rational person will do so. This must be the starting-point for all justified belief; if all beliefs
needed to be justified by other beliefs before they could be believed with justification, no belief could ever be justified.
All and only beliefs forced upon us by such inclinations are properly basic . 19 The reasons against going along with such
inclinations will come from other inclinations leading to contrary beliefs, which may be stronger inclinations and/ or
more numerous inclinations.

If it seems to me that I am seeing a table or hearing my friend's voice, I ought to believe this until evidence appears that
I have been deceived. If you say the contrary — never trust appearances until it is proved that they are reliable — you will
never have any beliefs at all.

Note the two very different uses of such verbs as ‘seems’, ‘appears’, and ‘looks’. When I look at a round coin from an angle I may say that ‘it looks round’ or I may say that
‘it looks elliptical’, but I mean very different things by the ‘looks’ in the two cases. By ‘it looks round’ in this context I mean that, on the basis of the way it looks, I am
inclined to believe that it is round. By ‘it looks elliptical’ in this context I mean that it looks the way elliptical things normally (that is, when viewed from above) look. The
former sense in philosophical terminology is the epistemic sense; the latter the comparative sense. The epistemic sense of such verbs describes how we are inclined to
believe that things are; the comparative sense describes the way things seem by comparing them with the way things normally seem. My statement of the Principle of
Credulity utilizes the epistemic sense.

That is, all basic beliefs are properly basic. Many philosophers have sought to develop more restricted principles governing the proper starting-points for rational belief, the
foundations on which a person's corpus of belief ought to be based. There is, for example, the principle which Plantinga has described and criticized so fully, called ‘classical
foundationalism’, that ‘a proposition p is properly basic for a person S if and only if p is either self-evident to S or incorrigible for S or evident to the senses for S ’. (See
A. Plantinga, ‘Reason and Belief in God’, in A. Plantinga and N. Wolterstorff, Faith and Rationality (University of Notre Dame Press, 1983), 59 .) But, as Plantinga has argued
at length, this is not a priori obvious: ought not beliefs about the mental which are not incorrigible and the deliverances of memory to be included? There is indeed a kind of
justification in which beliefs are not justified unless they start from (in some sense) objectively correct starting-points; and a system such as Classical Foundationalism might
tell us what these are. I shall come to discuss the notion of objective justification in Ch. 4 . But our only grounds for adopting such a system must be that it follows from
other things which we believe, and that means ultimately from beliefs which are properly basic on my internalist understanding. We have to start from where we find
ourselves, in respect of our beliefs as well as all else. It is because we cannot be faulted for where we start from that all basic beliefs are on this understanding properly basic.
But we can be epistemically at fault if (through not applying our own standards properly) we found the wrong beliefs on those starting-points — or do not eliminate some of
them because they are rendered improbable by others.

THE PROBLEM OF EVIL

For what would show that appearances were reliable, except more appearances? And if you cannot trust appearances
as such, you cannot trust these new ones either. But appearances may give grounds for believing some general view, a
consequence of which is that a particular ‘seeming’ should not be taken as showing how things are. If it seems to me
that I am seeing a table, but when I put my hand out it seemed to pass through where the table seemed to be, and so
did the hands of other people and so did other material bodies, then I ought to conclude that the initial ‘seeming’ was
delusory: perhaps what seemed to be a table was really a hologram. The Principle of Credulity applies not merely to the
deliverances of sense but to the way things seem morally, mathematically, or logically. If some inference seems to be
valid, I ought to believe that it is, until someone gives me good reason to believe otherwise. And it must apply also (in
the absence of reasons to make an exception in this case) to apparent experiences of God — if it seems to you that you
are aware of the presence of God, you ought so to believe (until someone produces reasons why you should be more
cautious in such a case).

It follows that, if it seems to someone that there is some bad state incompatible with the existence of God, he ought so
to believe, and so believe that there is no God — in the absence of counter-reasons. The incompatibility might involve
no further factual assumptions; namely, the inquirer may conclude that, whatever further factual assumptions you
make, the state is such that ( a ) or ( b ) or (d) are not satisfied. It might seem to him that the horror is so great that under
no circumstances would God have any right to let it occur, or that there is no sufficiently great good which it could
conceivably make possible which a God could not bring about in some better way. Alternatively, the inquirer might
believe that a God would have the right to bring about the bad state, and that it is necessary for a great good, but that if
there is a God, he (probably) does not do what else is needed to bring about that good. The inquirer might accept that
a God would be justified in bringing about certain bad states if he compensated the sufferer with life after death, but
claim that it is most unlikely that there is life after death. So again probably one of the justifying conditions — in this
case (c ) — is not satisfied with respect to the bad state. So, he must conclude, probably there is no God. All this, unless
he has evidence that seems to him to tell against that conclusion. The onus of proof has passed to the theist; he needs
reason for resisting the conclusion.

THE NEED FOR THEODICY

Many of us are surely often in this situation, and there would be something wrong with us morally if we were not. Our
understanding of an agent being good would be gravely deficient unless we thought that, other things being equal, a
good agent will stop pain and other suffering, if he can do so easily. And if our understanding of possible reasons why
anyone might allow suffering to occur provides us with no reason for supposing that a good God might allow certain
suffering, we ought to believe that there is no God — unless we have contrary reason. Just reflect on some of the
horrors that we read about in our newspapers and history books: the prolonged cruelty of parents to lonely children,
the torture of the innocent, the long-drawn-out acute physical pain of some disease, and so on. If we cannot see all that
as a reason for believing that there is no all-good and all-powerful being, when we cannot think of any reason why such
a being should allow it all to happen, there really is something deeply wrong with us. We have lost our sensitivity to the
good.

We may, however, have a contrary reason for not going along with this argument as far as to reach the conclusion that
there is no God. This contrary reason may be of three kinds. It may consist of other reasons for affirming that there is
a God; or general reasons for doubting the force of this sort of inference to his non-existence; or reasons for
supposing that criteria (a), (b), (V), and (d) are satisfied (a ‘greater good’ defence works) in respect of the particular bad
state which seemed initially to show the non-existence of God; in other words, a theodicy.

Let us begin with the first reason. No argument for denying a true condusion can be a sound deductive argument.
Either it is invalid, or it has one or more false premisses; and if it seems to a subject strongly enough (as a necessary
truth or a deliverance of experience) that some proposition is true, then any inference to its falsity must be judged
unsound. If it seems to a subject strongly enough that there is an external world or that 2 + 2 = 4, powerful
philosophical arguments to the contrary must be judged by the subject to be unsound, even if she cannot put her finger
on where they go wrong. Likewise, if it seems to a subject strongly enough (as a basic belief, delivered either by
experience or reason) that there is a God, then she is entided to the belief, whatever else she is initially inclined to
believe. In that case, the subject must conclude that any purportedly valid deductive inference from some bad state E
to the

THE PROBLEM OF EVIL

non-existence of God begins from a false premiss (E did not occur) or is not in fact a valid argument. If the argument
requires a further contingent premiss — e.g. that the conditions F under which God would be justified in allowing E do
not occur — and there is a purportedly strong inductive argument to this (i.e. an argument showing F to be
improbable), then (however improbable) either F holds, or E did not occur, or again the argument must be invalid. So
much is, I hope, undeniable.

But the basic belief will have to be overwhelmingly powerful if it is not to be overcome by an apparently simple
deductive inference, relying on apparently incontrovertible moral principles, from a very evident bad state. The less
sure the thinker is about the moral principles, the less strong the basic belief needs to be. Again, the less confident the
thinker is about the inferential principles, the less strong the basic belief in the existence of God needs to be in order to
overcome it. And finally, the weaker any inductive inference to not-F, the less strong the basic belief that there is a God
needs to be in order to outweigh it.

Even if the belief that there is a God is not a basic belief, as long as it seems to the subject to be supported by evident
basic beliefs on the basis of evidently cogent inferential principles , 20 it may be strong enough for the subject rationally
not to abandon it in the light of counter-argument. Again, it will depend on how evident are the basic beliefs and
inferential principles involved in the counter-argument and how probable it seems to the subject that they make the
existence of God.

However, in the absence of positive evidence of sufficient strength for the existence of God, we need some reason for
doubting the strength of the original atheistic inference in order to continue justifiably to believe. One kind of reason
will simply be a theodicy, grounds for supposing it is probable that the bad state in question is such that the four
criteria (a) to (d) are all satisfied in respect of it. The rest of this book will be concerned with developing a theodicy for
the various kinds of bad state which we find around us. As stated in the Introduction, that theodicy will assume that
God has done and will do for us the things which Christian revelation claims, e.g. that he will provide a life after death.
It will assume, that is, that any contingent justifying conditions required

For full discussion of the strength of arguments for the existence of God, see my The Existence of God, passin.

THE NEED FOR THEODICY

to satisfy (V), which are centrally affirmed by Christian revelation, are in fact satisfied. It will then proceed to argue that
the other conditions stated there are satisfied for each kind of bad state.

There exists, however, the possibility of a more general argument against the kind of inference involved in the atheist's
argument. I shall consider two such arguments. One is a very old argument to the effect that if there is a God, he will
be so much greater than us in his understanding of which states are good and which are bad that we should expect
there to be bad states for which we do not see how criteria (a) to (d) are satisfied in respect of them, i.e. in summary we
do not see what greater good they would serve. We do not know all the greater goods there are, and so we are not
entided to say what a perfecdy good God would allow to occur. This argument has recendy been deployed in a version
which has received quite a bit of attention, by Stephen Wykstra . 21

Wykstra recommends an amendment to my Principle of Credulity to obtain a principle which he calls CORNEA or
‘Condition of Reasonable Epistemic Access’:

On the basis of cognized situation s, human H is entitled to claim ‘It appears that ft only if it is reasonable for H to
believe that, given her cognitive faculties and the use she has made of them, if p were not the case, s would likely be
different than it is in some way discernible by her.

That is, we can only justifiably assert ‘It appears that ft if we have reason to believe that, if not-p, the observed situation
would be different in respect of those features which lead us to assert ‘It appears that ft. For example, we can only
justifiably say of some object ‘It appears that it is red’ if we have reason to believe that if it were not red, our experience
of it would be different. In general, if we are not colour-blind, we shall have such reason. But if we know that we are in
a room illumined only by a red light, we are not entitled to our claim. For we know that if the object were not red, our
experience of it would still be the same. Wykstra then uses this principle to claim that we cannot claim that it appears
that there are bad states which serve no ‘God-justifying purpose’, i.e. that it appears that there are states ‘such that God
(if he exists) would not allow them

21 Stephen J. Wykstra, ‘The Humean Obstacle to Evidential Arguments from Suffering: On Avoiding the Evils of “Appearance” Internal
16 (1984), 73-93 .

tional Journal for the Philosophy of Religion,

THE PROBLEM OF EVIL

to occur’. Wykstra claims that we cannot make this latter claim because we have no reason to think that if there were
no bad states such that God would not allow them to occur, we would recognize this fact; our moral beliefs being so
much shallower than those of God, if there is a God, we are likely to think some states of affairs to be bad states
serving no greater good, although in fact they do serve a greater good.

Wykstra's claim in his CORNEA is that we can only assert ‘It appears that ft if we have positive reason to think that if
p were not so, our experience would be different. But that is too bold a claim. The most Wykstra is entitled to by way
of amending my Principle of Credulity is to hold that we can only assert ‘It appears that ft if we have no reason to
think that if p were not so, our experience would be the same . 22 For if Wykstra's stronger claim were correct, what will
constitute a positive reason to think that if p were not so, our experience would be different? We may think that it
appears that we have such a reason (r). But before asserting ‘It appears that P, we must have reason to think that if r
were not so, our experience would be different. And so ad infinitum. If before asserting ‘It appears that it is red’ we
needed positive reason to think that if an object were not red, our experience would be different, we would indeed be
in a hopeless situation for making any justified assertions about the colour of the object. If we are ever to be justified in
making claims of the kind ‘It appears that ft, it must be the case that our inclinations to believe them suffice — in the
absence of counter-reasons. These latter will include any reason we have to suppose that if not-p, we would not notice
any difference. If we actually have a positive reason for supposing the room is illuminated by red light, then maybe we
are not entitled to the assertion ‘It appears red’; but only a positive reason could bar the justification. Given the basic
point that a positive reason for supposing that if p, we would not have any different experience, defeats any justification
we might have for concluding p on the basis of experience, the point is made more simply (and more fairly to ordinary
language) by saying

22 Wykstra defends this stronger interpretation against the weaker version in his follow-up paper, ‘Rowe's Noseeum Arguments from Evil’, in D. Howard-Snyder (ed.). The
Evidential Argument from Evil (Indiana University Press, 1996) . He accuses Rowe of misunderstanding CORNEA by interpreting it in the weaker way. See W. L. Rowe, ‘The
Empirical Argument from Evil’, in R. Audi and W. J. Wainwright (eds.). Rationality, Religious Relief and Moral Commitment (Cornell University Press, 1986) ; note his p. 237.

THE NEED FOR THEODICY

that the defeat occurs not by our no longer being entitled to say ‘It appears that ft but rather by our no longer being
entided to infer from ‘It appears that ft to p. If we have reason to suppose that the room is illuminated by red light, we
may still say that the object ‘appears red’, but we are not justified in inferring that it is red. That is, I can maintain my
Principle of Credulity in its original form, while allowing that among the reasons which make the inference no longer
correct (i.e. the conclusion no longer probable) is that things would appear the same to the subject even if they were in
fact quite different.

The only way to put the kind of point which Wykstra is trying to make in terms of my original Principle of Credulity is
to say that while it appears that there are bad states which serve no God-justifying purpose, i.e. ‘states such that God
(if he exists) would not allow them to occur’, we cannot infer that there are such states. This is because, Wykstra claims,
we have reason to suppose that if there is a God, his understanding of moral goodness will be so much deeper than
ours that it is to be expected that we should think that there are bad states which serve no greater good, when in fact
they do serve a greater good. But the trouble with this version of the argument is that while our moral beliefs (and
factual beliefs, we must add) may indeed be in error in relevant respects, we need some further argument to show that
they are more likely to be biased in the direction of failing to understand that some apparent bad states really serve
greater goods, rather than in the direction of failing to understand that some apparent good states really serve greater
bad states. Why should our inadequacies of moral belief lead us to suppose that the world is worse rather than better
than it really is? And if it is equally probable that the world is worse rather than better than it appears, it follows that it
is probably at least as bad as it appears; and very probable indeed (since there are very many apparently bad states
which apparently do not satisfy my criteria) that it contains at least one bad state which does not satisfy one of my
criteria [a) . . . (ft). True, if there is a God, then there will not be any apparent good states which really subserve greater
bad states. But we cannot assume at this stage without positive argument that there is a God and so that there are no
such states; for that would be to beg the question against the atheist who has produced the argument under discussion
in order to show that there is no God.

THE PROBLEM OF EVIL

Wykstra's version of the argument, that we cannot be expected to know all the greater goods for the sake of which a
God would allow bad states to occur, is only one of many slighdy different versions put forward recendy by theistic
philosophers . 23 Alston correcdy points out that God (if he exists) might allow some bad state of affairs for the sake of
some greater good, when we suppose falsely that he would not, either because the good state for which the bad state
was necessary was one of the nature and goodness of which we have no inkling; or because although we are aware of
the good state in question, we falsely suppose that it is logically 24 possible for an omnipotent God to bring about the
good without the bad. Both our moral intuitions and our logical intuitions may be in error. All that is indeed so. But
unless we have reason to suppose that our error is more likely to lie in the one direction rather than in the other,
Alston's point is worthless. What the Principle of Credulity tells us is that we must be guided by the way things seem to
us, in the absence of positive argument for supposing that certain appearances rather than others are misleading. And if
that positive argument depends on what God would do, it requires to be supported by argument to show that there is a
God, strong enough to oppose this argument from apparent bad states.

There is available to some inquirers, however, a more general argument against accepting the kind of inference
involved in the atheist's argument, which meets the problems involved in Wykstra's argument. It is an argument from a
premiss which, if true of the particular inquirer, would lead him reasonably to suppose that improvement in his moral
understanding would be in the direction of seeing all present puzzling bad states to be such as God would permit for
the sake of a greater good. This is that when initially it seemed to this inquirer that many kinds of bad states were
incompatible with the existence of God (directly, or given further contingent states of affairs which seemed to the
inquirer to be highly probable), subsequent reflection, argument, and experience led him to see, one by one, with
respect to a number of such kinds

See e.g. the two papers by Peter van Inwagen, the two papers by William Alston, and the paper by Daniel Howard-Snyder in Howard-Snyder (ed.). The 'Evidential Argument
from Evil .

Alston talks of ‘metaphysical’ rather than ‘logical’ possibility (ibid. 116—18). Whether or not we accept the distinction between ‘logical’ and ‘metaphysical’ possibility does not
affect the present issues, and so I phrase his point in terms of logical necessity.

THE NEED FOR THEODICY

of bad state, with apparent clarity, that they were not incompatible with the existence of God; a greater-good defence
apparendy worked in respect of them. But the reverse did not happen: the inquirer did not come to see states of affairs
which he regarded initially as good as apparendy incompatible with the existence of God (although he gave time to
considering whether there were any such states). It is satisfaction of the clause that ‘the reverse did not happen’ which
would provide an asymmetry of a kind not present in Wykstra's argument (without begging the question), necessary to
generate a conclusion favourable to theism. If what is hypostasized were true of some inquirer, that would give him
good grounds for supposing that eventually he would come to see with respect to all bad states for which initially a
greater-good defence could not be provided that such a defence could be provided.

Whether a given inquirer has this kind of evidence of finding more and more kinds of bad states explicable in terms of
a greater-good defence is an empirical matter. Certainly some inquirers have a record of such progressive
understanding; and an inquirer with such a record must rationally be much more doubtful about the moral principles
involved in his judging the remaining puzzling bad states to be incompatible with the existence of God than he would
otherwise be. Depending on the strength of his initial conviction, this evidence of his past record may tip the balance in
favour of God.

I conclude that it follows from the Principle of Credulity that bad states for which no greater-good defence can
apparently be provided must count against the existence of God. In order rationally to believe that there is a God,
despite this counter-evidence, we need either strong positive evidence for the existence of God, or a record of
discovering with respect to many apparent bad states that a theodicy works with respect to them, or a theodicy for each
kind of bad state which seems to count against the existence of God. The rest of this book (after the next chapter)
seeks, in the course of discussing the good purposes God might have for creating the world, to provide such a
theodicy. But first, in the next chapter, I need briefly to consider to what extent there are already in place in the
Christian tradition a number of strands of theodicy.

2 Theodicy in Christian Tradition

And unto Adam God said, Because thou hast hearkened unto the voice of thy wife, and hast eaten of the tree, of
which I commanded thee, saying, Thou shalt not eat of it: cursed is the ground for thy sake; in toil shalt thou eat
of it all the days of thy life; thorns also and thisdes shall it bring forth to thee; and thou shalt eat the herb of the
field; in the sweat of thy face shalt thou eat bread, till thou return unto the ground; for out of it wast thou taken;
for dust thou art, and unto dust shalt thou return.

(Genesis 3: 17-19)

The ‘problem of evil’ does not arise if one denies either the omnipotence or the perfect goodness of God (in the senses
of these terms delineated in the last chapter). Very occasional writers at different periods have denied the perfect
goodness of God, in the sense of his moral goodness — and we noted one modern example in the last chapter. But no
writer who could plausibly be called a Christian has denied the omnipotence of God until this century. In the early
centuries AD Manichaeism, a rival religion to Christianity, did hold that the reason why the bad occurs is that God is
not strong enough to prevent it; Good (God) and Evil were two powers equally strong in the world. In our century
many Christian theologians have talked in a vague way about the ‘weakness’ of God and seen this exemplified in the
crucifixion of Christ. But I do not think that many of them have really meant to claim that God (either the Father or
the Son, God incarnate) was too weak to prevent the crucifixion. Rather I suspect that talk about God being ‘weak’
was, for most of them, a somewhat misleading way of putting the point that God humbled himself in his incarnation in
Christ and did not exercise the power he possessed. However, one group of modern Christian writers who have clearly
and seriously expressed the view

THEODICY IN CHRISTIAN TRADITION

that God is not omnipotent and that the occurrence of the bad is explained by his inability to eliminate it are Process
Theologians. It is a feature of the system of theology that Charles Hartshorne, John Cobb, Schubert Ogden, and many
others developed from the philosophy of A. N. Whitehead, that God is limited in his power . 25

From time to time, Christian writers have expressed the view, mentioned earlier, that it would be logically impossible
for God to prevent the occurrence of all bad states. This view has usually derived from a view that the bad is simply
imperfection, the absence of some sort of good (privatio boni)\ and since any world is bound to lack some sort of
perfection, there will always be bad. But then the bad is not something real; it is, in some sense, nothing. This view
reached Christian thought from the Neoplatonism which exercised a powerful influence on it from the third to the
thirteenth century. Plato claimed that all the Forms and so all beings participate to some degree in the Form of the
Good. It follows that xd xoocov / malum (the bad, normally translated ‘evil’) is just inadequate participation, and so the
absence of some good. Pseudo-Dionysius wrote that ‘evil is contrary to progress, purpose, nature, cause, source, goal,
definition, will and substance. It is a defect, a deficiency, a weakness, a disproportion, a sin. It is purposeless, ugly,
lifeless, mindless, unreasonable, imperfect, unfounded, uncaused, indeterminate, unborn, inert, powerless, disordered.
It is errant, indefinite, dark, insubstantial, never in itself possessed of any existence .’ 26 But presumably the bad is not
just the absence of any good, but the absence of a good which in some sense ought to have been present. Writers in
this tradition clearly had in their mind the view that each thing belongs to a kind (participates in a Form essential to it)
and the bad consists in not being a perfect specimen of the kind. Thus Aquinas: ‘nothing that exists is called bad in so
far as it is an entity, but rather in so far as it lacks some being. Thus a man is called bad in so far as he lacks a virtue,
and an eye is said to be bad in so far as it lacks the power of sight .’ 27 This idea of the bad as in some way a lack of being
surfaces

25 See e.g. D. R. Griffin, ‘Creation out of Chaos and the Problem of Evil’, in S. T. Davis (ed.), Encountering Evil (T. & T. Clark, 1981) , where Griffin spells out how God's non-

26 De Divinis Nominibus, §32, trans. C. Luibheid (Paulist Press, 1987) .

27 St Thomas Aquinas, Summa Theologae , la. 5. 3 ad 2 .

THE PROBLEM OF EVIL

again in Hegelian thought, and in modern continental theologians influenced by Hegel or Heidegger. For Barth, evil is
das Nichtige, the ‘nothingness’ which inevitably accompanies the creation of creatures . 28

It does, however, seem very implausible to claim that pain and other suffering, bad desires, and wicked acts are just an
absence of some good — pain is not just an absence of pleasure, and wicked acts are not just the non-occurrence of
good acts . 29 It looks as if we can describe a world without pain or wickedness (even if it may lack certain other good
states). I suspect that most writers in the privatio boni tradition were half-aware of this, because they have usually
combined expression of this view with some view to the effect that God brought about the bad for the sake of some
good which could not otherwise be realized.

Most Christian thinkers have held (either with or more normally without the privatio boni view hovering in the
background) that God has allowed the bad to occur for some good purpose — whether or not we know what that
purpose is. That involves, given the later category of the logically impossible, the view that allowing the bad to occur is
logically necessary for the attainment of good. The view that some bad makes possible ‘greater good’ (as the point is
usually put) has been expounded by various analogies, especially aesthetic ones — a picture as a whole being more
beautiful in virtue of containing an ugly part, for example. ‘What horrifies us in a part by itself, may please us much
when we consider it as part of a whole,’ wrote Augustine. And ‘As the beauty of a picture is increased by well-managed
shadows, so, to the eye that has the skill to discern it, the universe is beautified even by sinners, though, considered by
themselves, their deformity is a sad blemish .’ 30 And, a millennium later, Leibniz also wrote ‘shadows enhance colours ’. 31

28 For references to Barth and analysis thereof, see Nicholas Wolterstorff, ‘Barth on Evil’, Faith and Philosophy, 13 (1996), 584—608 .

29 One of the last of the Scholastics, the 16th-century Spanish Jesuit Francisco Suarez, saw this clearly and tried to develop his own theory of the bad within the Scholastic
framework. See J. J. A. Gracia, ‘Evil and the Transcendentality of Goodness: Suarez's Solution to the Problem of Positive Evils’, in S. MacDonald (ed.), Being and Goodness
(Cornell University Press, 1991) .

30 Augustine, On True Religion, 40. 76 ; and City of God , 11. 23.

31 Theodicy , §12.

THEODICY IN CHRISTIAN TRADITION

The Free-Will Defence

Many Christian theologians have claimed not merely that God has some good purposes for the sake of which he
allows the bad to occur, but that such-and-such were God's actual purposes; or at any rate that such-and-such were
possible purposes to realize which God (omnipotent and perfectly good) could have allowed the bad to occur. In other
words, they have provided a theodicy for many of the world's bad states. And central to the view of almost all of them
has been the ‘free-will’ defence in respect of moral evil — that the bad states caused (or negligently allowed to occur) by
human beings result from their own free choice; and that it is such a good thing that humans should have free will that
it is worth the risk that they will abuse it in various ways. Almost all Christian theologians have affirmed that humans
have ‘free will’ — the only exceptions are a few of the classical Protestant theologians . 32 But there is reason to suppose
that some Christian theologians who affirmed ‘free will’ were affirming free will only in the compatibilist sense, and not
in the libertarian sense. By an agent having free will in the libertarian sense, to repeat, I mean that which intentional
action he does is not fully caused — either through some process of natural causation (i.e. in virtue of laws of nature) or
in some other way (e.g. by an agent such as God acting from outside the natural order). In that case whatever the
current state of the Universe (including the agent's beliefs and desires) and the causes at work in the Universe
(including those whose operation is codified in laws of nature), it remains possible either that the agent will do the
action in question, or that he will refrain from doing it. The compatibilist sense of ‘free will’ has been expounded in
various subtly different ways, but the basic idea is that someone has free will in this sense if they do what they want and
value (and do not act in consequence of psychological or physical pressure), even if they are fully caused to want and
value what they do . 33 The crucial difference for our

32 Luther wrote in his Disputation against Scholastic Theology (10): The will is not free to strive toward whatever is declared good. This is in opposition to Scotus and Gabriel
[Biel]’ (in Works, trans. H. J. Grimm, xxxi (Muhlenberg Press, 1957) ). For Luther, free will was confined to morally indifferent matters.

33 For some variants of compatibilist free will, see the papers of Frankfort and Watson in G. Watson (ed.), Fne Will (Oxford University Press, 1982); W. Neely, ‘Freedom and
Desire’, Philosophical Review, 83 (1974), 32-54; and D. Dennett, Elbow Room (Clarendon Press, 1984) . Of course libertarian free will belongs only to an agent choosing
intentionally between alternatives; it does not belong to any nonmental events, such as the random swervings of atoms, which are not fully caused.

THE PROBLEM OF EVIL

purposes is that an agent may have compatibilist free will even if God (or any other cause) causes him (via causing him
to want and value what he does) inevitably to make the choices he does, while he cannot have libertarian free will if
God (or any other cause) causes him inevitably to make the choices he does.

If the only free will humans have is compatibilist free will, there will be no distinction to be made between God
allowing some human to do a bad act, and causing him to do it. For then humans will inevitably do the acts they do
because of the way they are made. And if there is a God, it is God who made them that way. If they do bad acts, that
will be because God causes them to do bad acts. So if a compatibilist free-will defence is to work, what has to be
shown is that the actual occurrence of a bad choice when the agent wants and values what he chooses (not merely the
agent having the power to make a bad choice without being caused to do so) is a necessary condition of some good
state satisfying the demands stated earlier. And that is a more demanding requirement. The free-will defence clearly
becomes more plausible if Tree will’ means libertarian free will.

Just what various theologians meant by the Tree will’ they ascribed to humans is not always clear; although I think that
the context does often make it clear. But it seems to me that the natural primitive understanding of Tree will’ is as
libertarian free will. Our movements are not free if someone causes us to make them — my movement of my hand is
not free if someone else moves it for me. But if someone else (or something else) causes those movements, they are
not intentional actions of ours, i.e. caused by our wills. Analogously, our will is not free if someone or something else
causes us to will as we do — whatever the way that cause operates (e.g. whether via causing us to will what we value, or
to will what we do not value). To understand Tree will’ in a compatibilist sense seems to give the term a more
sophisticated, less natural sense. Hence, I suggest, the assumption must be (even if the context does not positively
suggest it), unless the context positively suggests otherwise, that when an ancient theologian ascribes to humans Tree
will’ and thereby seeks to excuse God from responsibility for our bad actions, he is ascribing libertarian free will.

THEODICY IN CHRISTIAN TRADITION

Bearing this point in mind, my assessment of the Christian theological tradition is that all Christian theologians of the
first four centuries believed in human free will in the libertarian sense, as did all subsequent Eastern Orthodox
theologians, and most Western Catholic theologians from Duns Scotus (in the fourteenth century) onwards. The main
Catholic declaration on this issue is that of the Council of Trent in the sixteenth century, which is firmly in favour of
human free will, and in my view the context of that declaration positively implies libertarian free will. St Augustine in
the fourth century seems to me, at any rate in his later writing, to believe that humans have only compatibilist free will.
Whether Western theologians subsequent to Augustine and before Duns Scotus did in general believe in libertarian
free will I am unclear; though I am clear that the greatest one, St Thomas Aquinas, was ambivalent on the issue . 34

In my view most of those who deployed the Tree will’ defence as part of their theodicy, to the extent to which they
were clear about what they were doing, were deploying a defence in terms of libertarian free will. The free-will defence
in this form is also a central plank of my own theodicy, and I shall be defending it at length over the course of later
chapters. I shall in future understand by Tree will’ free will in the libertarian sense, unless I state otherwise.

The Fall

The other strand central to many but by no means all theodicies has been the ‘Fall’ of the first human being, Adam;
and (less

34 For references to authors, in support of my assessment of the history, see my Responsibility and Atonement (Clarendon Press, 1989), 138-9. The authors referred to whom I
cite as believers in libertarian free will (e.g. theologians before the 4th century) often do not explain what they mean by ‘free will’, but since they are normally putting forward
the view that humans have ‘free will’ in contexts where they are seeking to show that God is not responsible for our bad choices or their consequences (e.g. the bad suffering
in Hell), this, together with my point about the natural primitive understanding of free will, means that we must understand them as advocating libertarian free will.Rowan
Greer has argued that St Augustine initially accepted the traditional view that humans have significant free will, but that in his later theology he regards only Adam and the
angels at the moment of their creation as having had this free will. (See Rowan A. Greer, ‘Augustine’s Transformation of the Free Will Defence’, Faith and Philosophy, 13
(1996), 471—86.) I gloss this transformation as one from a belief in libertarian free will to a belief merely in compatibilist free will (for all humans other than Adam).

THE PROBLEM OF EVIL

prominently) the Fall, before Adam, of angels, rational beings created by God with great powers, some of whom (their
leader often being called Satan or the Devil) chose the bad. Those theologians who, like Augustine in his later period,
came to believe that all later humans had at most compatibilist free will were prepared to ascribe to Adam and the
angels (before their respective falls) libertarian free will, and regarded much, if not all, of the later moral and natural
evil in the world as due entirely to their (libertarianly) free bad choices. The more numerous theologians who held that
all humans have some free will (and in my view that means libertarian free will) saw the choices of Adam and the bad
angels as having a significant (though limited) role in causing natural evils, and in bolstering the bad desires of later
humans (though not predetermining whether or not they would yield to those desires).

Bad angels and in particular Satan are mentioned occasionally in the Old and New Testaments, and both the Book of
Job and the Revelation of St John picture them as having significant influence in the world for bad. Later theologians
developed the theory that the angels at the beginning had just one free choice, which would fix their subsequent
character for ever. If an angel chose a good character, that was fixed for ever; and if the angel chose a bad character,
that too was fixed for ever. Angels, good and bad, were given temporary influence over the world. Church tradition
saw Satan as the serpent of Genesis 3: 1, who tempted Eve, who in turn tempted Adam to commit the first human
sin. 35

But theology has given much greater prominence to Adam, the first human, as the source of much of the world's bad
states. This is because of the explicit statement in the opening book of the Bible, that human suffering and death were
due to the sentence passed by God on Adam and Eve for their sins. 36 Later theology did not take the details of this
passage (the temptation being to take a fruit etc.) very seriously. Likewise it skipped over the fact that, according to
Genesis, the first human sin seems to be that of Eve rather than that of Adam. But it did in general affirm that there
was a first human sin, call it the sin of Adam, from which severe consequences flowed. St Paul, in a famous passage
which

“Man sinned at the suggestion of The Devil,’ decla
Gen. 3: 16-19.

Fourth Lateran Council (M. Denzinger, Enchiridion Symbolorum , 23rd edn. (Freiberg, 1963),

800 ),

THEODICY IN CHRISTIAN TRADITION

takes up the Genesis story, wrote that ‘through one man sin entered the world and death through sin; and so death
passed unto all men, for that all sinned ’. 37 St Paul does not here mention Adam's sin as the cause of later suffering, only
of death and sin; but many later Christian theologians have seen it as the cause of later suffering.

However, the view that the sin of Adam had a role in causing the death of later humans is a view somewhat more
central to the Christian tradition than the view that Adam's sin had a role in causing the suffering of later humans. I
shall be arguing later that death is not, as such, a bad state; it is simply the absence of a good state — although it may be
closely connected with bad states of suffering in the dying or those close to them.

If this claim that the sin of the first human caused the subsequent proneness to suffering and death of later humans is
read as the claim that the first human was naturally immortal and not prone to suffering, and that his (i.e. Adam's) sins
changed that nature, it seems a claim very unlikely to be true in view of the close similarity of the nature of humans to
that of animals. It seemed very unlikely even to many ancient thinkers who knew nothing of our common ancestry
with the animals. Mortality and suffering seemed to them to be involved in the embodiedness of an organism as such,
quite apart from any bad choice that it or an ancestor might have made. Hence the claim that Adam's sin caused the
subsequent proneness to sin and death has often been read as the claim that while Adam was born liable to suffering
and death, if he had not sinned, God would have intervened in the natural order to preserve him and any descendants
from suffering (from which in his privileged position in Eden he had so far been preserved) and from death. And
given the Christian doctrine that, in consequence of Christ's resurrection, God will give everlasting life (of some kind)
to all (or at any rate to the good), it seems natural to suppose that if humans had not sinned and so needed Christ's
redeeming work to restore them to wholeness, everlasting life (in this world or another one) would have been theirs
anyway. That the effect of Adam's sin was to stop God acting supernaturally to prevent

Rom. 5: 12. Ron
St Paul gives no

, 5: 12—21 was the source used by Augustine for his doc
warrant for that doctrine in this passage.

ter). I have argued ( Responsibility and Atone?.

THE PROBLEM OF EVIL

suffering and death was the view of Augustine , 38 Aquinas , 39 and Scotus .'" 1

Augustine and almost all who followed him in Western Christendom until the last two centuries held that Adam's sin
caused (directly, or in the indirect way elucidated alone) not merely our suffering and death, but our sinfulness, i.e. the
bad desires or inclinations, the proneness to do wrong both of later generations and of Adam himself, who did not
(before his sin) suffer from this proneness. This proneness — as a major part of what theology has called ‘Original
Sin’ — subsequent Christian theology has normally claimed, was transmitted by nature, not nurture, that is by some
genetic means rather than by bad example . 41 Even if this claim is read (analogously to the preceding claims) as the claim
that Adam had a suppressed sinfulness, while being given supernaturally a temporary non-proneness to sin; and that
but for Adam's sin, God would have intervened to bestow a similar supernatural gift on subsequent humans, the claim
is open to an initial difficulty not raised by the other claims: how, if he had no active bad desires, could he have sinned?
I shall be arguing later that we cannot knowingly do wrong in the absence of bad desires. And Genesis emphasizes, as
did subsequent theologians, that Adam sinned knowingly. There must have been the smallest of inclinations to sin
already in him before it was abundantly magnified by Adam's subsequent actual sin . 42

38 Augustine writes of the ‘privileged condition’ of Adam and Eve in Paradise as ‘a condition mysteriously maintained by nourishment from the tree of life, which would have
been able to preserve them from sickness and from the ageing process’ (De Genesi ad Utteram , 11. 32). That is, they needed external help to be preserved from mortality.
Aquinas understands Adam's immortality to amount to the fact that God gave him a special supernatural grace which would have kept him from death, but which was
withdrawn when Adam sinned ( Summa Theologae , la. 97. 1 and 2).

40 Quaestiones in Lib. ii sent. , dist. 29. 5. See N. P. Williams, The Ideas of the Fall and of Original Sin (Longmans, Green, 1927), 409-10 .

41 See the Council of Trent (Denzinger 1513 and 1523) for its firm affirmation of the mechanism of transmission of ‘Original Sin’. Original sin so transmitted included both
sinfulness and some sort of responsibility for doing something about the consequences. This ‘responsibility’ has been understood by some as guilt, and so a bad state. I give
in the text very brief reasons for rejecting that understanding. In my view the responsibility is a responsibility to deal with the consequences of a sin for which we later
humans are not guilty. For justification of this understanding, see my Responsibility and Atonement , 145—6 .

42 Augustine distinguished the state of the Blessed in Heaven involving their inability to sin ( non posse peccare ) from the state of Adam, who had the ability not to sin (posse non
peccare ), but the latter did involve, according to Augustine, a certain openness to temptation — which it seems impossible to construe except as a very mild form of
sinfulness. (See Opus Impefectum c. lulianum , 5. 61 and City of God , 12. 8.) Yet Augustine clearly did not want to put the matter in that way; and to impose consistency on
anything Augustine has to say about the moral condition of human beings is a hard task.

THEODICY IN CHRISTIAN TRADITION

But Augustine had a rather different view of Adam from that of most of the theologians who preceded him, as well as
of theologians in the East later than Augustine. Irenaeus wrote that ‘man was a child, not yet having his understanding
perfected. Wherefore he was easily led astray by the deceiver .’ 43 That is, Adam was created as a weak creature and so
already significantly prone to sin. If (non-suppressed) sinfulness was part of Adam's created nature, it follows that
(barring divine intervention) we would have inherited sinfulness whether or not Adam himself had sinned.

For theologians before the fourth century, the ‘Fall’ was a fall from a lower pedestal than it was for later Western
theologians. But beginning with Athanasius 44 a quite different doctrine of Adam developed mainly in the West, as a
perfect man without significant proneness to sin, a doctrine of Adam's ‘original righteousness’ which reached its
extravagant fullness in Augustine. He held that Adam was ‘exempted from all physical evils, and endowed with
immortal youth and health which could not be touched by the taint of sickness or the creeping debility of old age ’. 45 If
we had inherited that nature , 46 we would have had little proneness to sin; and so

Demonstration , 12. See also his A.dversus Haereses , 4. 38. 1. In the Eastern Orthodox tradition, a thousand years later St Gregory Palamas writes of ‘our ancestors’ ‘being still
in an imperfect and intermediate state — that is to say, easily influenced, whether for good or evil’. It was because of their immaturity that it was not good for them to eat of
the tree of the knowledge of good and evil, which, however you interpret it, claimed St Gregory, clearly provided a superior kind of knowledge. Our ancestors ought instead
first ‘to have acquired more practice and so to speak, schooling in simple, genuine goodness’ ( Topics in Natural and Theological Science , in G. E. H. Palmer, P. Sherrard, and K.
Ware (ed. and trans.), Philokalia , iv (Faber & Faber, 1995), 369-70 ).

Contra Gentes , 2. 4.

See Williams, The Ideas of the Fall and of Original Sin , 361, for references to Augustine's views on this.

This doctrine of Adam's original righteousness was taken up in the Orthodox Confession of Peter of Moghila (as approved by the Council of Jassy, 1642), one of the
important doctrinal statements of the Orthodox Church (though not one with authority similar to that of the Seven Ecumenical Councils). It claimed that Adam's innocence
was joined ‘with every perfection and innate righteousness, both on the side of intellect and on the side of will’. Williams (The Ideas of the Fall and of Original Sin , 543), whose
translation I quote, comments that ‘the definitely Augustinian tone’ of these sections ‘contrasts strongly with the vagueness which had previously characterised Eastern
thought on this subject’.

THE PROBLEM OF EVIL

Adam's sin was seen as the cause of our sinfulness. That view became normal in the West. By contrast an important
doctrinal statement of the Orthodox Church (the confession of Dositheus, ratified by the Council of Jerusalem in
1672) seems to support the view that the consequences of Adam's Fall were confined to suffering and death. 47

Even if one accepts that Adam's sin caused our sinfulness, sinfulness is only a proneness to sin; and, given that
libertarian free will remains, we can resist this proneness. If libertarian free will does not remain, then of course it is
predetermined whether or not we will yield to temptation. Put another way, all theologians in the West subsequent to
Augustine held, against Pelagius, that we need (in consequence of our sinfulness) God's gracious help — his ‘grace’ — in
order not to sin when temptation occurs. But those who believed that we still have libertarian free will believed that we
then have the free choice of whether to use this grace or not. Those who did not believe that we have this free will
believed that if sufficient grace was not provided we would, if tempted, inevitably sin; and that if sufficient grace was
provided we would inevitably not sin.

Augustine claimed that, as well as causing suffering and death for humans and our inclination to sin, Adam's sin made
us guilty for that sin. That view was in general unknown to earlier theologians, except to several important ones who
explicitly denied it. 48 Augustine, however, fastened that doctrine on the Western Church for many years, until the later
medievals greatly toned it down or abandoned it. Classical Protestantism temporarily reverted to the

‘We believe that the first man was created by God and fell in Paradise, when, neglecting the divine command, he obeyed the deceitful counsel of the serpent; and that thence
has flowed in succession the ancestral sin so that no one is born according to the flesh who does not bear this burden and perceive the fruits thereof in this present world.
And by the “fruits” and the “burden” we mean, not sin, such as impiety, blasphemy, murder . . . etc.: for many of the Patriarchs and Prophets, and myriads of others of those
both who lived in the shadow of the Law and in the truth of the Gospel, and the divine Forerunner, and especially the mother of God the Word, the ever-virgin Mary, had
no share in these or like sins; but we mean those things which the divine justice sends upon man as a penalty on account of the transgression, such as the sweat of toil,
tribulations, bodily infirmities, the pains of child-birth, a toilsome life during this pilgrimage, and finally bodily death’ (Decretum 6, trans. Williams, The Ideas of the Fall and
Original Sin, 542) .Much turns on how ‘had no share in’ is read. But the Decretum does not mention increased proneness to sin, let alone guilt, as among the consequences of
the Fall.

48 For references and fuller discussion of the doctrine of Original Guilt, see Responsibility and Atonement , 144—6.

THEODICY IN CHRISTIAN TRADITION

Augustinian doctrine. The Eastern Orthodox tradition has had little sympathy with this doctrine. I shall not discuss it
further. It is clearly to be rejected on moral grounds: I cannot be guilty in respect of the sins of another.

As I wrote earlier, the Fall of Adam has played a role of varying prominence in the theologies of different theologians.
While any theologian had to take account of the Genesis story, one could see it merely as describing an incident in early
history without enormous cosmic significance. Alternatively, a few of the early Fathers thought of it not as history but
as a representation by a myth of the state of each one of us. Gregory of Nyssa writes of Moses, the supposed author of
Genesis, as ‘placing doctrines before us in the form of a story ’. 49 And that has been the view of many theologians of the
past two or three centuries. In that case there is no doctrine at all of a historical Fall causing later bad states. I shall wish
to accept a historical Fall, and give it some role in my theodicy, but not the kind of prominence which Augustine gave
to it. In any case, as most theologians have seen, it could not solve the ‘problem of evil’ fully, it merely takes it one
stage further back. There would still remain the problem why God allowed the moral evil of Adam's sin; or, if Satan
had a role in this, the moral evil of Satan's sin — though it may be that, in this form, the problem is more readily
soluble. And it is important to note that, although the Fall of Adam (and to a lesser extent of the angels) has had some
importance in Christian theology, neither of them gets any explicit mention in Christian creeds and that not even the
Fall of Adam has been seen as a doctrine right at the centre of the Christian faith.

Other Strands of Theodicy

The Old Testament contains several meditations, subsequent to that of the Book of Genesis, on theodicy, the best-
known of which is that of the Book of Job. But exactly what is the final conclusion of that enigmatic book is much
disputed. In any case it is a pre-Christian conclusion. The New Testament contains no systematic treatment of the
source of all the world's bad states, though St

THE PROBLEM OF EVIL

Paul's Epistle to the Romans contains a lengthy treatment of one particular bad state — the rejection of Jesus Christ by
the Jews. The only clear conclusion of this section of the Epistle is that this bad state is only temporary and allowed to
occur only for the sake of a great future good which it makes possible. And that is the theme prominent elsewhere in
the New Testament with respect to other bad states. The parable of the tares and the wheat 50 emphasizes that the tares
will be allowed to grow only until the harvest time, then they will be bound in bundles and burnt. But if the tares are
dug up before the harvest, there is the risk that the wheat will be dug up with them before it is ripe. Jesus's reply to the
question of his disciples (in St John's Gospel) about the man born blind, ‘Who did sin, this man or his parents, that he
should be born blind?’, was: *N either did this man sin nor his parents, but [it happened in order] that the works of God
should be made manifest in him .’ 51 The explanation of the world's bad states lies not in their past causes but in the
future good which they make possible.

Very many Christian writers have stressed the value of suffering for the human beings who suffer, in enabling them to
form their souls for good. By showing courage and sympathy in the face of their suffering and that of others, people
can become naturally good people. That is a theme especially prominent in the Eastern Orthodox tradition, which I
shall warmly endorse. Then there has been the claim that the bad of much human sin is outweighed by the great good
of the Incarnation resulting in Christ suffering with us 52 and above all redeeming us on the cross. While acknowledging
that much bad is indeed mitigated by these subsequent good states, I shall deny that the subsequent good can possibly
outweigh the bad which led to it. Another theme often deployed in Christian thinking is that some suffering is
punishment for the sins of those who suffer. That is possible, but no major Christian theologian has claimed that this
explains all suffering: how could it possibly explain the suffering of babies before they are capable of sin? (And note the
words of Jesus cited at the end of the last paragraph.)

John 9: 1—3. ‘Nor his parents’ is clearly to be read in a sense which rules out also more remote ancestors.

In recent theology Jurgen Moltmann (The Crucified God \ trans. R. A. Wilson and J. Bowden (SCM Press, 1974) ) has emphasized the great good for us, that God in Christ
manifests the love which suffers in solidarity with those who suffer. He does not, however, regard this as providing a full solution to the problem of evil.

THEODICY IN CHRISTIAN TRADITION

The Christian tradition, while deeply worried about moral evil, has been less worried about natural evil, and less
worried still about the problem of animal suffering. During most of Christian history, natural evil was often seen as an
inevitable consequence of finitude, in the form of being embodied and mortal (the latter in the case of humans being a
consequence of the Fall). For embodiment involves being at the mercy of the natural world: fire can burn us and water
can drown us. And embodiment involves ‘corruption’; our bodies decay, are subject to disease, and become weak and
worn out. But it seems to me logically possible to have disease, even incurable disease, without agonizing pain; and
those who write in this way do not give any rigorous arguments to show that this is not possible. Augustine, among
many others before and since, pointed out that the natural causes of suffering can be used for good purposes, if we
choose to use them for these purposes: ‘Poisons which are destructive when used injudiciously, become wholesome
and medicinal when used in conformity with their qualities and design; just as, on the other hand, those things which
give us pleasure such as food, drink and the light of the sun, are found to be hurtful when used immoderately or
unreasonably .’ 53 True, but it was surely not beyond the power of God to make medicines which could not poison, and
food in which we felt no inclination to indulge immoderately.

This same unthinking assumption about the inevitability of suffering for creatures embodied in a natural world has
continued in a lot of modern theology. Schleiermacher, who may be considered the founder of the tradition of modern
continental theology, derived this from the interdependence of the nature in which we are embodied: ‘The very same
activity or condition of a thing by which it enters on the one hand into human life as an evil, on the other hand is a
cause of good, so that if we could remove the source of life's difficulties the conditions of life's progress too would
disappear .’ 54 F. R. Tennant has argued with somewhat greater persuasiveness that we need to live in an orderly world (i.
e. one governed largely by laws of nature) if we are to learn and make choices; and that an orderly world is inevitably
one in which natural evil occurs. So ‘if water is to have the various properties in virtue of which it

53 City of God , 11. 22.

54 F. Schleiermacher, The Christian Faith, sect. 48 (Eng. t

trans. ed. H. R. Mackir

id J. S. Stewart (T. & T. Clark, 1989), 187) .

THE PROBLEM OF EVIL

plays its beneficial part in the economy of the physical world and the life of mankind, it cannot at the same time lack its
obnoxious capacity to drown us ’. 55 That may be so if water and humans are made as they are. But it seems to me far
from evident that there could not be a different system of chemistry (deriving from a different system of physical laws),
such that there was a substance which played the beneficial role of water without embodied rational creatures being at
risk of death by drowning in it.

I think, nevertheless, that despite these evident difficulties, there is something right in the claim that embodiment and
finitude, in which creatures have an ability to learn, involve natural processes producing suffering as well as good. One
of my tasks in this book will be to show in just which qualified form this claim is correct.

So I shall want to use many of the various strands in the Christian tradition of theodicy, emphasizing some and
downplaying others, in the articulation of my own theodicy. But what I shall also use from the Christian tradition is its
understanding of what is good and bad for the inanimate world and for human and animal life. In listing the good and
bad states in Part II, I shall bring out there and in my final chapter that the insights on this of the Christian tradition
(most of which could be attained without religious assumptions, by reflective secular thought) provide the materials for
a satisfactory theodicy. I shall want to stress not only the value of human free will, but above all the value for us of our
lives being of use to others. There have in the Christian tradition also been many writers who affirm that since we
know that there is a God, we know that there is some true theodicy (a true account of possible, and indeed actual,
reasons why God would allow moral and natural evil to occur), but deny that we have any idea what it is. In
communities where the apparendy knowledgeable people (whom we have reason to trust on theoretical matters) tell us
that God exists; where the apparently knowledgeable people can produce arguments from the natural world or from
scripture purporting to show that God exists; where some people purport to have direct awareness of God; and where
there are no counter-arguments in circulation, a belief that there is a God will be strongly justified. It will be so well
justified as to overwhelm any apparent objections from the bad states of the world, so that people

Philosophical Theology, ii (Cambridge University Press, 1930),

THEODICY IN CHRISTIAN TRADITION

in those communities will all have a well-justified belief that God has good reason for allowing the bad states to
occur — even if they have not the slightest idea what that reason could be. Many in the past were in that situation, but
many of us today are not; and hence the need for theodicy or at least progress towards one. And so I move to set out
my theodicy. Part II will describe the good purposes God may be expected to have (and the Christian revelation claims
that he does have). Part III will show that if God seeks to realize these, he inevitably allows the bad states of this world
to occur; but that (given the Christian revelation) he does all else he can to secure the good. Part IV will argue that God
has the right to allow the bad, and that the good outweighs the bad.

This page intentionally left blank

II The Good Goals of Creation

This page intentionally left blank

3 Beauty

In the beginning God created the heaven and the earth. And the earth was waste and void; and darkness was upon the
face of the deep: and the spirit of God moved upon the face of the waters. And God said, Let there be light: and there
was light. And God saw the light, that it was good: and God divided the light from the darkness. And God called the
light Day, and the darkness he called Night. And there was evening and there was morning, one day. And God said,
Let there be a firmament in the midst of the waters, and let it divide the waters from the waters. And God made the
firmament, and divided the waters which were under the firmament from the waters which were above the firmament:
and it was so. And God called the firmament Heaven. And there was evening and there was morning, a second day.
And God said, Let the waters under the heaven be gathered together unto one place, and let the dry land appear: and it
was so. And God called the dry land Earth; and the gathering together of the waters called he Seas: and God saw that
it was good. And God said, Let the earth put forth grass, herb yielding seed, and fruit tree bearing fruit after its kind,
wherein is the seed thereof, upon the earth: and it was so. And the earth brought forth grass, herb yielding seed after its
kind, and tree bearing fruit, wherein is the seed thereof, after its kind: and God saw that it was good. And there was
evening and there was morning, a third day. And God said, Let there be lights in the firmament of the heaven to divide
the day from the night; and let them be for signs and for seasons, and for days and years: and let them be for lights in
the firmament of the heaven to give light upon the earth: and it was so. And God made the two great lights; the greater
light to rule the day, and the lesser light to rule the night: he made the stars also. And God set them in the firmament of
the heaven to give light upon the earth and to rule over the day and over the night, and to divide the light from the
darkness: and

THE GOOD GOALS OF CREATION

God saw that it was good. And there was evening and there was morning, a fourth day.

(Genesis 1: 1-19)

A perfecdy good God will seek to do many good actions and no bad ones. Good actions often derive their goodness
from bringing about states of affairs which are intrinsically good, i.e. good because of what they are and not because of
how they were brought about or what they cause. Although it is not always, as we have seen, a bad action to bring
about an intrinsically bad state of affairs, it requires the special conditions described in Chapter 1 to make it not a bad
action. Other things being equal, a good God will seek to bring about much good and no bad. I plan in Part II to list
many of the good states which a perfectly good God might seek to bring about, and many of the bad states which
(barring special conditions) he will not bring about. I shall point out also that Christianity has claimed, with respect to
many of these good states, that God does indeed have the purpose of bringing them about. The goodness of good
states and the badness of bad states arise from their being good or bad for someone or something; and every state
which is good (or bad) for someone or something is (in that respect) good (or bad). We shall need to be clear about
who or what the person or thing is which is being benefited or harmed. This is because, as I shall argue in Chapter 12,
if God is to have the right to allow some bad to come about, that can only be if God gives to the individual to whom
the bad happens a life in which on balance the good outweighs the bad.

I have no general formula for picking out good (and bad) states of affairs; I believe them to be too diverse to fall under
any such formula. That that is so was one of the main conclusions of chapter 6 of G. E. Moore's Principia Ethica. There
is, he wrote, ‘a vast variety’ of ‘great intrinsic goods’ and ‘great intrinsic evils’. Another main conclusion of that chapter
is that most of them are ‘highly complex wholes composed of parts which have little or no value in themselves’. 56 While
wishing to emphasize that the goodness or

G. E. Moore, Prinicipia Ethica (Cambridge University Press, 1903), 224 . Moore held, however, that while ‘evils’ could be necessary constituents of great goods, in fact the
‘evils’ of this world always outweigh the goods which they make possible. Hence (p. 220), ‘we cannot admit the actual validity of any of the arguments commonly used in
theodicies’.

BEAUTY

badness of complex wholes is often very different from that resulting from any arithmetical sum of the goodness or
badness of parts, I demur from Moore's suggestion that many ordinary things have no value in themselves. Surely
every concrete thing — stick or stone — has value in itself; it is intrinsically good that it exists. And since this goodness
does not arise solely from it being good for someone or something else that it exists, we should say (odd though it may
sound initially) that it is good for it that it exists.

The existence of all concrete things — of stones and icebergs, of wind and fire, and mountains and valleys — is good in
itself. The more of them, the better. And better that they be arranged in a beautiful way. Could anyone who has come
to admire sculpture possibly deny that? But better still is a moving sculpture — a process whereby trillions of trillions of
concrete things emerge from simple beginnings. Could anyone who has come to admire dance possibly deny that? And
good that they should come in kinds with marvellous patterns of colour, new kinds emerging from old — a living
painting. The goodness of the existence and beauty of the non-conscious world — both the simpler inorganic world and
the more complex world of plants, of trees and grasses and flowers — is so obvious; and yet it needs a poet to bring it
alive. Some of the prose poetry of the more popular science books does quite well in bringing alive the simplicity of the
principles of behaviour (laws of nature) which governed the behaviour of the immensely hot and dense quanta of
matter-energy at the time of the Big Bang; how they cooled to form the chemical elements and then larger molecules;
how the vast clouds of stuff coalesced into stars which formed galaxies, and cooler bodies — planets; and how the
larger molecules came together to form simple and then more complex plants. My poor pen is ill equipped to depict
the greatness of the dance of the Universe. But is it not obvious that a good God would seek to bring about such
beauty?

The Bible begins with the affirmation cited above (expressed in the form of a hymn, representing the process of
creation as taking place in idealized day-stages) that God did bring about the inanimate world, because it was good in
itself. And this has been a constant theme of subsequent Christian theology. All the medievals thought that everything
that exists (i.e. every concrete thing that exists) is good in virtue of that very fact that it exists. They disagreed among
themselves about whether that goodness was

THE GOOD GOALS OF CREATION

intrinsic to anything existent; or whether it belonged to created things in virtue of their being created by, and so in
some sense ‘participating in’, a good God. I suggest that we take the former position, which was that of Augustine : 57
the goodness of stones is not dependent on who created them; God created stones because for other reasons (i.e.
intrinsically) their existence is good.

It may be objected that there is nothing good in a work of art if no one observes it. This objection, which is of a type
which we shall meet several times in this book, seems to me mistaken. When people admire a beautiful work of art,
they are not admiring the effect of the work on their own consciousness; they do not say ‘This is a wonderful painting
because it produces these feelings in me and other people who look at it 5 . Rather, it is because they believe that it has
some beautiful feature, which they have been fortunate to notice, that it does produce the appropriate feelings. The
‘feeling’ involves as an essential part the belief that the painting is beautiful in itself, that it has a beauty which it would
still have even if no one had the right sensory equipment to notice it. Of course it is also good that people admire what
is beautiful; but the beauty of the beautiful does not depend on being recognized. How could it? For recognition of
beauty, as of anything else, depends on the existence of the feature before and independently of being recognized.

But anyway, even if I were mistaken about this and there is only a point in the existence of beautiful things if someone
admires them, God has a very good reason for making a beautiful Universe, namely that he himself will admire it (not
admire it because he made it, of course; but because what he made is admirable). And of course it is good that the
beautiful Universe also be admired by very many other conscious beings; that is a good for them. That is a further
point in creating it. And at the end of the twentieth century we humans on Earth are in the privileged position of being
able to look through our telescopes and see it, not merely as it is now, but as it was in the early stages of its emergence
from the Big Bang.

While the existence of any concrete thing is a good, its non-existence is not a bad state of affairs; it is just the absence
of a good state. If the non-existence of things was intrinsically bad, then any

57 ‘Whatsoever things are, are good’ (Augustine, Confessions , 7. 12). For analysis of medieval disci;
Goodness (Cornell University Press, 1991), introd.

: things are good, see Scott MacDonald (ed.), Being and

BEAUTY

world would be a very bad place indeed — because there would be so many things it did not contain. If one thinks of
ugliness as a negative quality, as opposed to being the mere absence of beauty, one would be hard put to think of any
part of the pre-human world which is ugly; ugliness in this sense seems to arrive with the arrival of humans, who,
knowingly or unknowingly, make something which could be beautiful ugly instead.

The goodness of the inanimate world (in its very existence and its powerful beauty) and so (if there is a God) God's
interest in creating it are too obvious to need further prose discussion. ‘Consider the lilies of the field,’ said Jesus, ‘how
they grow; they toil not, neither do they spin: yet I say to you that even Solomon in all his glory was not arrayed like
one of these .’ 58

Matt. 6: 28-9.

4 Thought and Feeling

And God said, Let the waters bring forth abundandy the moving creature that hath life, and let fowl fly above
the earth in the open firmament of heaven. And God created the sea-monsters, and every living creature that
moveth, which the waters brought forth abundantly, after their kinds, and every winged fowl after its kind: and
God saw that it was good. And God blessed them, saying, Be fruitful, and multiply, and fill the waters in the
seas, and let fowl multiply in the earth. And there was evening and there was morning, a fifth day. And God said,
Let the earth bring forth the living creature after its kind, cattle, and creeping thing, and beast of the earth after
its kind: and it was so. And God made the beast of the earth after its kind, and the cattle after their kind, and
every thing that creepeth upon the ground after its kind: and God saw that it was good.

(Genesis 1: 20-5)

Good though the inanimate world is, clearly the animate world of animals and humans is even better, and a major
reason for this is that animals and humans have mental lives 59 of belief and thought, desire, sensation, and purpose. I
shall consider in this chapter the involuntary mental states, and in particular beliefs and desires — which are states with
which subjects find themselves, not through their own choice and which they cannot change in the short term (though
they can take steps which will lead to change of these in the

I understand by a ‘mental’ state one to which the subject has privileged access, one about which — if she so chooses — she can know better than can anyone else. Mental states
include conscious states, such as sensations (e.g. an image of blue in my visual field) of which we must to some extent be aware while we have them; and states such as beliefs
which continue while we are not aware of them. On the nature and taxonomy of the mental life, see my The Evolution of the Soul , ' rev. edn. (Clarendon Press, 1997), pt. 1 .

THOUGHT AND FEELING

long term); and go on in the next chapter to consider purposes manifested in intentional action. I seek to list the many
good such states which a perfecdy good God might seek to bring about, and to contrast them with the bad states
which (barring special conditions) he will not bring about.

Beliefs

Beliefs are views about how the world is. We find ourselves with certain beliefs — that I am in Oxford, that today is
Monday, that there is no greatest prime number — which we cannot change in the short term. We can, however,
investigate them further and then come to believe what our investigations reveal (which may or may not be different
from what we believed to start with). (Or we can sometimes brainwash ourselves over time into having some belief
which we desire to have; and then try to forget that we have done this.) But a ‘belief’ would not be a belief if we could
change it readily at will (and so realize that we are doing it). For if some attitude is to be a belief we must think of it as
corresponding to how things are in the world and so as liable to change as things change in the world and thus as
caused by the world. If we could choose our ‘beliefs’, we would have no reason to suppose that they corresponded to
how things are, and thus we would not regard them as beliefs . 60

True belief is a very basic good. Many true beliefs are of course instrumentally good. A true belief about which road
leads home will enable me to bring about the fulfilment of my desire to be at home. But my suggestion is that true
beliefs are also intrinsically good (both for the believer and for the objects of belief) — all true beliefs, but especially
ones which concern rational agents (e.g. humans) and other sentient beings (e.g. animals), and the history of the
environment in which they have lived. One of the greatest misfortunes that can happen to a person, the ancients
realized, is for his ‘memory’ (i.e. memory about him) to be ‘blotted out ’. 61 And even today we recognize it as a great
misfortune for someone who is dead when he is falsely believed to have done something wrong: hence

60 For fuller justification of the involuntary

61 See e.g. Dante, Inferno , III. iv. 9.

character of all beliefs.

ibid. 126-8.

THE GOOD GOALS OF CREATION

public campaigns to ‘clear’ someone's ‘name’. This is because what people did and what happened to them matters;
and recollecting these things consciously (and so having them available in accessible memory or written record) gives
them a certain life after death (in a metaphorical sense) and constitutes an acknowledgement of their importance and a
tribute to them. It is good for those who do things and to whom things happen that they be ‘preserved’ — truly; a false
memory does not acknowledge the actual deeds and happenings.

True beliefs about past individuals are especially worth having where what is believed is to the credit of those
individuals, e.g. that they did great things, despite much suffering . 62 But so long as there is some good in their lives, they
are worth remembering; and remembering the bad things as well fleshes out the memory of the good things, and
makes the whole person ‘come alive’. And mere living — the existence of an animate being which itself has
beliefs — means that there is some good in that life.

As it is a good for the past individuals to be recalled, and as it is a good for anyone to have the privilege of benefiting
others (a point I shall wish to emphasize a lot more in due course), so it is a good for present individuals to have the
privilege of recalling past individuals — of having true beliefs about individuals worth recalling, and of being the
vehicles of telling others about them. The greater the connection of the past individuals with ourselves, the more
appropriate it is that it is we who recall them. And even the inanimate merits respect: it is a good for the Universe that
we have a true belief about its constitution and history, and in so far as it has formed us, it becomes more appropriate
that we pay that tribute to it.

These things are also a good for us, not only because we give a certain life after death to the past, but because we
acquire thereby, also in a metaphorical sense, a certain control over them. To understand is already to begin to master.
Control over things is a good for us (a point which also I shall need to emphasize a lot more) — like so many goods,
capable of abuse, but a good all the same. The goodness

62 The Wisdom literature of the Old Testament is a very powerful advocate of the value for those who achieve worthwhile things, of being recalled. ‘In the memory of virtue is
immortality, because it is recognized both before God and before men’ (Wisd. 4: 1). The famous passage from Ecclus. 44, so often read at ‘Commemorations of
Benefactors’, contrasts the blessedness of those who ‘have left a name behind them’ (v. 8) with the wretchedness of those who ‘have no memorial’ (v. 9).

THOUGHT AND FEELING

of true beliefs on important matters constitutes, I think, the major reason why benefactors give large sums of money
for research into the distribution of the galaxies, the ‘initial state’ of the Universe, the history of science and culture. Of
course such research, like research into the life stories of our immediate relatives, may have some instrumental value.
But it is implausible to suppose that many of those who finance and pursue it do so primarily because of that
possibility. They clearly regard the acquisition of true beliefs on these matters as intrinsic goods, and I have tried to
explain what makes them such.

Especially important is to have true beliefs about moral matters — to believe that it is good to give to the starving and
bad to torture children. Obviously there are instrumental reasons why true beliefs here matter: they guide our actions.
And the ‘control’ reason also applies. But there is clearly a further reason here: we are intrinsically better people for
having the right outlook on moral matters. Our goodness is in part a matter of how we view the world; someone with
false moral views is the worse for that, even if he is in no position to act on them.

It is good that our true beliefs should be strong ones: they guide our actions more firmly and enable us to see more
clearly how the world is. Is strong true belief the better if it amounts to knowledge, if it has also that extra
something — let us call it ‘warrant’ — which turns it into knowledge? That depends on what ‘warrant’ is. My view is that
‘knowledge’ is a fairly vague notion and that there are different ways of spelling it out, some of them largely internalist,
some more or less externalist. An internalist theory of knowledge claims that the ‘warrant’ which turns a true belief
into knowledge arises (in major part) from something internal to the subject (i.e. something to which she has ready
access). A belief has warrant, for example (to put the point loosely, at this stage), if it is supported by the subject's other
beliefs, or is based on immediate experience. An externalist theory claims that warrant arises solely from something
external to the subject to which she may have no access. The most common version of externalism is reliabilism, which
holds that a belief has warrant when it is produced by a process such as perception which normally produces true
beliefs . 63

63 For an extended statement of reliabilism, the most common version of externalism, see A. Goldman, Epistemology and Cognition (Harvard University Press, 1986), pt. 1 . For a
traditional statement of intemalism, see Roderick Chisholm, Theory of Knowledge, 3rd edn. (Prentice-Hall, 1989) . For an introduction to epistemology, contrasting internalist
and externalist theories, see Keith Lehrer, Theory of Knowledge (Routledge, 1990) . A full statement of the proper functioning version of externalism is to be found in Alvin
Plantinga, Warrant and Proper Function (Oxford University Press, 1993) .

THE GOOD GOALS OF CREATION

Now clearly it is a good thing that our beliefs satisfy the reliabilist requirement, for the fact that they do means that
(given the evidence of their method of production) they will probably be true. But, if a given belief of mine is true, I
cannot see that it is any more worth having for satisfying the reliabilist requirement. So long as the belief is true, the
fact that the process which produced it usually produces true beliefs does not seem to make that belief any more worth
having: just as so long as a desk is beautiful and useful, it is no better for having been produced in a factory that
normally produces beautiful and useful products rather than in one which normally produces ugly and useless
products. An alternative form of externalism spells out ‘warrant’ in terms of ‘proper functioning’: a belief has warrant
if it arises in us as a result of the proper functioning of our cognitive apparatus. If there is no God, then proper
functioning will amount to (roughly) functioning the way Evolution (in some sense) planned us to function; but I
cannot see why a true belief would be the better for functioning in this way. If there is a God, then proper functioning
amounts (roughly) to ‘functioning the way God intended it to function’. This sort of proper functioning might well add
to the worth of a true belief and that would be because thereby it constituted the fulfilment of a desire or purpose, in
this case God's desire or purpose — on the goodness of the fulfilment of which I shall be commenting in due course.

What about ‘warrant’ in an internalist sense? The core component of ‘warrant’ for an internalist is ‘justification’. To
amount to knowledge, a (strong) true belief must be ‘justified’. But there are different ways of understanding
‘justification’, according to whether the justification is a matter of satisfying the subject's own immediately accessible
criteria or whether it is a matter of satisfying the true criteria, which a subject might not have initially in a clear form but
which he can reach by reflection. The true criteria are the necessary a priori inductive criteria of what are the proper
starting-points for belief, and of what makes what probable. The a priori is internally accessible; that is, what is true in
this field does

THOUGHT AND FEELING

not depend on how things are in the world outside ourselves but solely on the relations between concepts which we
can discover by inner mental reflection. Yet the a priori is not necessarily what we believe it to be on first thinking
about it. This can be illustrated by mathematical beliefs, which provide a paradigm example of beliefs which can
amount to a priori knowledge. I might initially think that 5 + 7 = 13 or that there is a greatest prime number; reflection
will lead me to see that 5 + 7 = 12, not 13, and that there is no greatest prime number. Similarly, with our inductive
criteria I might think initially that the fact that a coin has landed ninety-nine ‘heads’ in a row after one ‘tail’ makes it
probable that it will land tails next time (the ‘Monte Carlo’ fallacy), but later realize that that is not so: rather the fact
that it has landed heads ninety-nine times out of 100 is evidence that the coin is biased and so will land heads next time.

I considered the subjective internalist sense of justification in Chapter 1, because our concern there was with subjects
having beliefs which they were not epistemically at fault in having. In that sense a belief is justified if (by the subject's
own criteria) it is rendered probable by the subject's other justified beliefs, or it is properly basic. Following an earlier
terminology of mine, 64 let us call such beliefs ‘justified ’ or ‘rational ’. But while a subject may not be at fault in
operating in accordance with her own criteria, clearly it is better if she operates by true criteria: a belief is justified in an
objective sense either if it is rendered probable by the subject's other justified beliefs or if it is properly basic. A belief
thus justified I shall call ‘justified^ or ‘rationaiy The difference from the previous definition arises from the omission
of the words ‘by the subject's own criteria’ which govern both of the following clauses. We all believe that there are true
criteria for what makes what probable, for what is evidence for what. To hold that e makes h probable is to hold that it
does so by true criteria; we believe that the criteria each of us applies in making such judgements are the true ones;
otherwise we would regard it as equally permissible to hold the wildest beliefs on the grounds of present evidence — to
think, for example, that ninety-nine heads in a row makes it probable that five blue parrots will fly into the room. In
fact, not merely do we hold that there are true criteria for what makes what probable and that we have these, but that it
is also the case that almost everyone has very

See my Faith and Reason (Clarendon Press, 1981), 45 .

THE GOOD GOALS OF CREATION

similar criteria and so a very similar view about what are the true criteria. I need, for later purposes, to summarize
briefly what (as almost all will agree) are some of these true criteria . 65 (Note that I use the word ‘likely’ with the same
meaning as ‘probable’; and sometimes write elsewhere of the probability of something happening as the chance of it
happening.)

First, a hypothesis is more likely to be true, in so far as it renders probable what we observe (e.g. gives true
predictions). Secondly, the probability given to a hypothesis by the fact that it makes it probable that we will observe
what we do is diminished in so far as those observations are ones which are fairly probable anyway (that is, even if the
hypothesis in question is false). Thirdly, a hypothesis is more likely to be true in so far as it is simple. It is an important
consequence of the second criterion that a hypothesis of the form ‘All As are B’ is more likely to be true if we observe
many As which are B (and none which are not B) than if we observe only a few. For the more we observe, the less
probable it is that we will find what we observe (namely, more and more As all being B) if it is not the case that all As
are B. For the same reason such a hypothesis is more likely to be true if the As which we observe are observed under
varied conditions; if all As which we observe are observed in certain narrow conditions, C, it is almost as probable that
there is no general regularity that all As are B, but only the regularity that all As under conditions C are B. And, finally,
what As and more varied As, all found to be B, show is not a coincidental connection between being A and being B,
but a physically necessary connection — most simply that being A causes something to be B, or, less simply, that being
A and being B have a common cause . 66

65 If you do not believe that there are these true criteria of probability, then among the wild criteria which there is no reason not to adopt are criteria from which it follows that
the fact that other people are screaming is no reason for supposing that they are in pain; and that (even if she cannot recall so asking) anyone's suffering is always suffering
which she has asked God to give her. With criteria like this, we would conclude that there is very litde suffering, apart from what we have asked God to give us; and the
problem of evil would be solved very quickly. The problem is a serious problem only because we assume that our criteria of probability are correct ones.

66 I am assuming that tA ’ and ‘B ’ are projectible predicates, e.g. predicates like ‘emerald’ or ‘green’, rather than predicates defined in complicated ways in terms of such
predicates, such as ‘grue’ (where an object is said to be grue at a time if and only if the object is green and the time is before AD 2000 or blue and the time is after AD
2000). It will not be appropriate to discuss here the important issue of how projectible predicates are to be recognized as such. On the latter, see e.g. Nelson Goodman, Fact,
Fiction and Forecast , 2nd edn. (Bobbs-Merrill, 1965), ch. 4 ; and my An Introduction to Confirmation Theory (Methuen, 1973), ch. 7 .

THOUGHT AND FEELING

Other important evidential criteria which I believe to be a priori include criteria for inferring to people's thoughts and
feelings from the way they behave, and to their beliefs from what they say; and a criterion to the effect that (in the
absence of counter-evidence) we should believe what people tell us about what they themselves believe. For people are
aware of what they believe; if we do not trust what they tell us about their beliefs without further positive evidence of
their reliability, we would have no grounds for most of our beliefs about the lives of those close to us or about history
or geography etc. We have beliefs about these matters because others have told us what they believe, and that they have
their beliefs because they are rendered probable by others of their beliefs. I believe what I do about history because
others have told me (in writing) what they believe and that their beliefs are based on certain documents which they
believe that they have seen. If I could not trust what others write about history without checking them out, I would
have very little by way of justified historical beliefs. All of these criteria are criteria of what is evidence for what — that
is, of which of our beliefs (about what we have seen or experienced, or others have told us) make other beliefs
probable.

We also have our a priori criteria, and believe that they are in general true criteria, of proper basicality — that is, of
which beliefs have some degree of probability simply on the basis of it seeming to the subject that they are true;
criteria, that is, for what sort of content a belief has to have in order to be a correct starting-point for the process of
forming other beliefs. Clearly any belief which is necessarily false (e.g. that 2 + 2 = 5) is not a properly basic belief. And
although perhaps even in this objective internalist sense it is right for us to start from whatever beliefs (about
contingent matters) we find ourselves with, a belief about the apparent deliverance of perception forms a better
justified starting-point for belief-building than some general world-view (e.g. the truth of quantum theory) or some
claim about the future (e.g. that it will rain tomorrow) of which we are convinced. There is an important reason for this
which we can reach by reflection. Our beliefs about what we are currently perceiving are caused by a short causal chain
from a state of the public world to which those beliefs purport to correspond. For the beliefs to be true, only that short
chain has to operate

THE GOOD GOALS OF CREATION

correctly. But if we have a belief about what will happen tomorrow or which scientific theory is true, there can be no
direct causation (given the logical impossibility of the future causing the past). Any mechanism correlating belief and
state believed will be indirect, in the sense that it will depend on two separate causal chains (some earlier unobserved
state causing both the belief and the state believed). Such a mechanism will be more liable to incorrect operation and so
to producing false beliefs than one where the belief concerns what we currently observe. A basic belief that the sun
rose today will inevitably be a more properly basic belief (i.e. have a much greater probability, be a better justified
starting-point for belief) than a basic belief that it will rise tomorrow. For the same reason a belief about what I am
currendy observing is as such a more properly basic belief than a memory belief about what I saw yesterday (where the
causal chain is much longer).

All of these criteria — both of what is evidence for what and of proper basicality — are a priori. They are part of the
armoury with which we come to experience and in virtue of which we can extrapolate from it to the past, the future,
the unobserved, and very general theories. Unless we had a priori criteria we could take no step beyond experience.
(We do of course use other rules as well for judging what makes what probable and what are proper starting-points for
belief, which are empirically based, e.g. that this kind of witness often lies, or that fingerprints at a place make it
probable that the one and only person with prints of that shape was present at that place. But we use these rules
because it can be shown probable that they yield correct results by a priori properly basic beliefs in virtue of a priori
criteria of probability.) Scientists would never agree — as they almost always do — about which theory their evidence
supports unless they had these common criteria, and we would regard their results as arbitrary unless we thought that
in general their common criteria were the correct ones.

Now again it is good that our beliefs should be justified^ because it matters that we have true beliefs, and if they are
justified^ they are probably true. But the best any of us can do towards having justified^ beliefs at a given time is to have
justified beliefs, beliefs which seem to us at that time to be probable on our criteria (i.e. the ones which seem to us
correct). Over time we can try to improve our criteria. But because it is the best we can achieve at a time, it matters

THOUGHT AND FEELING

also that we should at each time have justified beliefs. The internal coherence of our beliefs gives us in a different
important respect the good of a rational outlook on the world. However, if knowledge is understood in an internalist
sense as involving ‘justified true (strong) belief’, clearly ‘justified ’ is the more plausible candidate for the justification
involved. Someone who by a lucky accident had strong beliefs which are true and rendered probable by his own crazy
standards could not be said to have knowledge. The person who claims to know that the coin will land tails because
ninety-nine of the 100 past tosses have been heads does not really know that — even if the coin does land tails.

In order to have an account of knowledge in anything like the ordinary sense, the internalist needs to add two further
conditions to the account so far of knowledge as justified true (strong) belief. First, he needs a condition to deal with
the ‘Gettier effect’, that if the justification for a true belief ‘proceeds via’ some false belief, then we do not have
knowledge. 67 Clearly it is a good thing that this condition be fulfilled also, if only for the reason that it is good that all
our beliefs should be true. An internalist will also need to add a condition to the effect that to amount to knowledge a
belief must be ‘based on’ its evidence. That is, if what justifies the belief B is that it is rendered probable by other beliefs
of the subject, these other beliefs must cause the subject to hold B . 68 The objective relations of probability must be
those which affect the subject in her belief formation. Even if the subject's belief B is made probable by her other
beliefs, it must not be a mere coincidence that this is so; it

67 The simplest kind of example to illustrate the case where someone has a justified true belief the justification for which proceeds via a false proposition is as follows. I believe
with justification that you own a Ford car (because I have seen you drive one). But if I am justified in believing that you own a Ford car, I am justified in believing any
proposition which that entails, for example that ‘Either you own a Ford car or you will win the lottery’. Suppose I do believe this latter proposition. You do not, however,
own a Ford, but you do win the lottery. So I have a justified true belief that either you own a Ford car or you will win the lottery. But plausibly this does not amount to
knowledge, because my justification for believing it proceeds via a false proposition — that you own a Ford. This kind of example was drawn to philosophical attention by
Edmund L. Gettier, ‘Is Justified True Belief Knowledge?’, Analysis, 23 (1963), 121—3 . For discussion of the kind of extra condition needed, see Keith Lehrer, Knowledge
(Clarendon Press, 1974), ch. 1 .

68 The ‘basing’ clause may need to be phrased more carefully than this, but the main point of such a clause should be clear. For more sophisticated accounts of this, see K. A.
Korcz, ‘Recent Work on the Basing Relation’, American Philosophical Quarterly 34 (1997), 171-91 .

THE GOOD GOALS OF CREATION

must not be the case that B is formed by wishful thinking and the subject has not seen the force of evidence in favour
of B constituted by her other beliefs.

Now if knowledge is construed as justified 2 strong true belief based on its evidence (that is, on our properly basic
beliefs) (where the justification does not proceed by a false proposition — the Gettier condition), is it a better thing to
have than is mere strong true belief (with a Gettier condition satisfied)? If we respond to our evidence in accordance
with correct criteria for assessing it, it is more probable that we shall get true beliefs than if we assess it in accordance
with incorrect criteria. But why should it matter if an individual true belief has been formed in accordance with
incorrect criteria? I claimed earlier that it would not matter if an individual true belief was not formed by a reliable
process. But there is a difference here, arising from the fact that forming beliefs in accordance with criteria is normally
a process of which the subject (at any rate the average human subject) is almost always at least half-conscious. We are
at least half-conscious of the criteria we use in inferring from evidence to hypothesis (though it may need a philosopher
or psychologist to make them fully explicit). But then, one who operates on false criteria will have false beliefs about
what is evidence for what, about the right way to assess beliefs. And false beliefs here are false beliefs about something
very central to human rationality. So if true belief matters, true belief which is justified is even more valuable.
Furthermore, we value the great scientists of the past not just for getting the answer right, and not just for having true
beliefs about justification, but for being influenced by those beliefs in forming their scientific theories; for seeing the
force of complex evidence and being guided by it; for responding consciously to the world in their beliefs in the right
way. A person who does this does not merely have rational beliefs in an important sense, but is a rational being. I
conclude that knowledge in the internalist sense is a better thing to have than mere strong true belief; though
knowledge in the most common externalist sense, the reliabilist sense, is not. A capacity for knowledge in the objective
internalist sense is one of the hallmarks of human beings.

The Christian tradition has always seen its capacity for having knowledge of deep truths, with the allied capacity to
reach such knowledge by reason (see next chapter), as one of the great-making features of humans. The Greek
background of the tradition of

THOUGHT AND FEELING

course stressed this. Plato emphasized the difference between ‘true opinion’ and ‘reason’, holding that ‘only the gods
and a small group of humans partake of reason ’. 69 Aristode's Metaphysics, however, begins with the claims that ‘all
humans by their nature desire to know . . . Animals other than humans live by sense, impression and memory with but
a small share in connected experience, whereas the human race lives by art and science .’ 70 For both writers the
knowledge that mattered was knowledge of the underlying principles of things. For all Christian theologians from the
beginning of the Christian tradition, man was made in the image of God; and a primary element of that image was the
power of reasoning. What makes man different from the animals which were created before him is his intellectual
nature. John of Damascus said about this what all others also said: God ‘creates with his own hands man of a visible
nature and an invisible, after his own image and likeness: on the one hand man's body he formed of earth, and on the
other his reasoning and thinking soul ’. 71 In stressing its connection with the reasoning power of humans, theologians
make it clear that the knowledge which they see as so good is knowledge in the internalist sense.

But it is not only human beings who have knowledge. The higher animals know things about their environment: where
the food or predators are, where their young are, how to feed the young, and so on. These are not mere beliefs. They
are true beliefs, either properly basic ones or ones formed on the basis of experience, namely on the basis of other
beliefs which make them probable (though of course animals do not have an explicit concept of probability). And the
other conditions for knowledge are also normally satisfied. It is good that higher animals act in the light of knowledge
rather than respond instinctively as do the lower animals. It is good that birds get worms, because they know that their
young want them and they seek to feed the young. It is good that the deer avoids the fire, because it knows that the fire
is dangerous. If knowledge is the good we believe it to be in humans, a lesser degree and amount of knowledge in
animals is a lesser good.

Reason is shown not merely in the possession of beliefs which amount to knowledge, but in responding to new
evidence with changing beliefs. Learning from experience (as opposed to being

70 Metaphysics 980 a b ; my italics.

71 De Fide Orthodoxa , 2. 12.

THE GOOD GOALS OF CREATION

told by some expert) is, I suggest, a further good beyond the mere possession of knowledge. We value the great
scientists not merely for having beliefs based on evidence in the right way, but for changing and developing their beliefs
as new evidence turns up, for learning from experience. Consider some scientist Smith who originally held (with
justification , given his evidence) some quite false theory of the Universe T ; and then, being shown new evidence with
which that theory was not compatible, and suddenly thinking of a new theory T supported by all the evidence new and
old, comes to hold the new theory. Contrast him with Jones, who was shown in his youth all the evidence which Smith
eventually saw and, told about T , comes to hold it because of the evidence. Both are equally rational in their response
to evidence, but Smith alone in a crucial sense learns from experience. He alone manifests a rationality prepared to give
up old views in the light of new evidence. When we look at the history of science, we see that our heroes are not as
such those who hold true theories on good grounds, but those prepared to abandon old theories in the light of new
evidence, in order to adopt the true theories on good grounds. We value the process of knowledge acquisition in the
light of new evidence additional to the product, knowledge.

I began this section by dwelling on the goodness of true belief. By contrast all the reasons that make true belief good
make false belief bad. To believe that people have done and suffered things that they have not done and suffered is not
merely not to acknowledge them as they are, but is clearly worse than that. I have already commented on the bad state
of falsely believing someone to have done wrong. But it is also a bad thing (though a less bad thing) to believe someone
A to have done good when they have not. It means that we regard them in the same way as someone B of whom we
believe truly that they have done some similar good — and that is an injustice to B, for it involves failing to recognize
what is special about B. It means too that our relation to A depends on a falsehood and so that any true beliefs about
the good that is in A become distorted because we do not have a proper picture of A.

THOUGHT AND FEELING

Desires

I understand by a ‘desire’ or ‘want’ an involuntary inclination with which an agent finds himself to do some action or
to have something happen . 72 Desires may be for almost anything — for mental states, including sensory states, for
bodily states, for states of the world far distant from the agent. I may desire to have a certain tingling sensation or for a
piece of poetry to run through my mind, to waggle my ears, or to be President of the United States. Or my desires may
be focused on others — that my children be happy or successful or inherit my wealth when I am dead. The satisfaction
of desire is an evident good for the desirer (unless the desire is for something bad). It is good that a conscious being
who wants something shall have that want satisfied. It is good for it that its longings for how the world should be
should be respected by the world. The satisfaction of desire is also a very basic good in the sense, as we shall see, that
its goodness contributes to the greater goodness of many other states of affairs.

Enjoyment or pleasure consists in the believed satisfaction of present desire; and that is also an obvious good (for its
subject) unless the desire is for something bad. It normally comes on top of the actual satisfaction of desire and
therefore is an additional good. Pleasure then includes doing what you are inclined to do, or letting happen what you
are inclined to let happen, when you believe truly that you are doing the action or letting the state occur. I enjoy eating
cake or playing golf if I am inclined to do so, do so while inclined, and believe that I am doing so. I get pleasure out of
sitting in the sun if I let myself continue to sit in the sun when inclined to do so (do not struggle against sitting there)
and believe that that is what I am doing. I get pleasure out of being President of the United States if I am inclined to be
President, and am, and believe that I am President. I stress that pleasure consists in the believed satisfaction of present
desire. This may be a desire which I have had for a long time but which has been until now (I believe) unsatisfied; or it
may be a desire which came to me simultaneously with it being (I believe) satisfied.

r a fuller account of desire, see my

.For

The Evolution of the Soul, ch. 6 .

THE GOOD GOALS OF CREATION

The most primitive kind of pleasure is sensory pleasure, pleasure in the having of certain sensations, such as the
occurrence of a warm feeling in my throat. I do not see any reason to suppose that there are any sensations which are
intrinsically pleasant or intrinsically unpleasant. For consider any sensations which most humans dislike, say, the
sensation of burning. That sensation can occur in a very weak form in which we do not dislike it (we may even like it),
and in some exceptional people in a somewhat stronger form in which they do not dislike it. Masochism is a genuine
phenomenon. What seems rather to be the case is that humans (and animals) are (almost) all so made that they like
(desire to have) the same sensations (to approximately the same degree) and dislike (desire not to have) the same
sensations. To call a sensation a ‘pain’ would seem to imply not merely that it is of a certain kind, but also that it is
disliked. Also, not all disliked sensations are naturally called ‘pains’ — a disliked feeling of nausea is not naturally called a
pain. The occurrence of any disliked sensation, whether we call it ‘pain’ or not, is a bad state of affairs (for the one who
has the sensation); and the occurrence of any liked sensation is a good state of affairs (for the one who has the
sensation).

Conscious states such as sensations are states of affairs about the occurrence of which I can hardly be mistaken.
However, the more remote from my immediate environment is the state of affairs on which my inclinations are
focused, the more serious is the possibility that I may believe some inclination to be satisfied when it is not. I may
believe that I am President of the United States when I am not. I still get pleasure out of this, for pleasure consists
merely in the believed satisfaction of desire, but there is not here also the good of the actual satisfaction; and when I
am disabused of my false belief, there is nothing left out of which I can get pleasure.

Pleasure is the believed satisfaction of present desire. Is there good in the present satisfaction of past desire? I suggest
so, if the desire is unabandoned in the sense that while the agent still has desires, he still has that one. Suppose that,
aged 5, 1 long to be an engine-driver but I cease to have this desire when I am 10, when I desire to be a naval captain
instead; if I am then declared medically unfit for the Navy when conscription arrives and am drafted to become an
engine-driver instead, there would seem litde (if any) good in this. But this is because my earlier desire has been
replaced by a different desire. Yet contrast this with the desire of a dying

THOUGHT AND FEELING

man for being buried here rather than there when he is dead. That desire is not abandoned; it is the man's last wish.
And in that case surely it is good for him that his relatives should bury him here rather than there, even if but for the
dead man's known past desire, there would be nothing particularly good about burial here rather than there. The
reason why it is good does not arise solely from any consequences that it may have in leading other people to believe
that they too will be buried where they desire and get pleasure out of that belief. For the goodness of the satisfaction of
the dead man's desire would remain even if a law is just about to be passed making cremation obligatory in future; we
would say then of the dead man ‘Thank goodness he, at any rate, got what he wanted’. The reason is simply that it is
good that longings be satisfied; they matter, and that could not be unless it matters that their goal is reached.

Both the satisfaction of desire and the believed satisfaction of desire are good things (for the desirer), unless the desire
is for what is bad. Conversely, the non-satisfaction of desire is a bad thing (for the desirer), again, unless the desire is
for what is bad. (A desire is not satisfied if it is not fulfilled at the time to which the desire relates. A desire to go to
London tomorrow is today neither satisfied nor non-satisfied.) I suggest that non-satisfaction is bad even if the desirer
does not believe that the desire is unsatisfied. The non-satisfaction of my desire to be buried here rather than there is a
bad thing for me. Suffering — to use a more general word than pain for this bad state — is the believed non-satisfaction
of desire. That is a bad state, even more evidently. For therein consists unhappiness. The believed non-satisfaction of
desire is a bad thing even if in fact the relevant desire is satisfied. It is bad if I believe that my desire to be in California
is not satisfied when in fact it is. In such a case there are good and bad states to be weighed against each other. But of
course things are worse if the desire which the subject believes not to be satisfied is in fact not satisfied. Such non-
satisfaction I call ‘frustration’ of desire. Sensory pains are one species of such frustrations. They involve two
components: a sensation and a strong desire that it cease (which I shall call an aversion to it), known not to be satisfied.
But an unfulfilled longing to be President of the United States (known to be unfulfilled) is a bad thing for just the same
sort of reason . 73

73 See Additional Note 2 .

THE GOOD GOALS OF CREATION

The goodness of a desire is, however, affected by what it is a desire for. The satisfaction (or believed satisfaction) of
desires for things intrinsically bad is itself a bad thing. The satisfaction of a desire that others, or even one's later self,
suffer pain, or lose their reputation, fortune, or family is of course a bad thing (for them); and any pleasure derived
from these things is not merely not a good, but very much a bad state — for the one who is pleased . 74 Only the pervert
rejoices at the sufferings of others. Also, I suggest, pleasure is not on balance a good where the belief needed to sustain
it is false. The pleasure which a man gets from believing his son to be a successful businessman when in fact he is
unemployed is not a good thing for the father. We can see this when we consider that if we had the opportunity to plug
into an ‘experience machine’ which would inculcate in us the false beliefs that our desires were currently being satisfied,
we would — almost all of us — refrain, under normal circumstances (i.e. unless life without the machine was so
intolerable that plugging in was the lesser of two evils). While, I am claiming, the satisfaction of any desire for anything
(so long as it is not a bad state) is good, the satisfaction of a desire is the better if it is a desire for a state of affairs good
for other reasons. Desires to read great novels rather than low-quality thrillers, to understand quantum physics or to
develop a correct theodicy rather than know Wisden's Cricketers' Almanack by heart are like this, because of the width
and depth of the knowledge we desire to attain in the former cases . 75 Since the satisfaction of desire is as such a good
thing, it

The situation is different when the desire is not for a state of affairs bad in itself but for one associated with a bad state of affairs. If my desire to win the lottery being
satisfied involves the frustration of your desire to win the lottery, but is not in itself a desire for the latter, then the satisfaction of my desire is still a good. But whether the
total state of the satisfaction of my desire and the frustration of yours is overall a good depends on how great are the good and bad components respectively
J. S. Mill implausibly suggested that the fact that those who are ‘capable of the higher pleasures’ often choose them showed that the higher pleasures were more pleasurable
(see his Utilitarianism (1861), ed. M. Warnock (Collins, 1962), ch. 2 ). But that of course does not follow at all, unless you already assume that the only things which move to
action are desires to have pleasure and avoid pain; a theory even more dubious than Mill's ethical theory. The point is surely that some things are more worth enjoying than
others; and even if by some measure there is a greater quantity of pleasure to be had out of the latter, it is (up to a point no doubt) better to have the former. Mill's famous
sentence ‘it is better to be a human being dissatisfied than a pig satisfied; better to be Socrates dissatisfied than a fool satisfied’ (p. 260) is true. And, as Mill says, Socrates or
the human being who know ‘both sides’ can see that But they can also see, as Mill omits to note, that the pig or the fool may well (if it could be measured) get more
pleasure — but those who see both sides do not think that that in itself makes a difference to the relative goodness of the two states.

THOUGHT AND FEELING

must be that the satisfaction of a strong desire is as such a greater good than the satisfaction of a weak desire.

The satisfaction of joint desires — e.g. the desires of two creatures for the common end of their nest being built — is
very good (for both creatures) because of the goodness of sharing and cooperation. And it is yet better that the desire
of one for the satisfaction of the desire of another — e.g. my desire for the satisfaction of your desire to eat cream cake,
or more generally the desire of a mother for the satisfaction of her offspring's desire — be satisfied, than that the desire
of each of us for the satisfaction of his or her own desire be satisfied. This is because of the goodness of mutual
concern and involvement. Much better that my desire that your desire to eat cream cake be satisfied than that the
desire of each of us for the satisfaction of his or her own desire be satisfied. (All this, given that the desires are not for
what is bad.) It follows, because of the goodness of mutual concern, that even better is the satisfaction of joint desires
for the satisfaction of the desire of a third creature, e.g. the satisfaction of the desire of both parents for the desired
success in examinations of their child. Even better still is the satisfaction of desires to perform actions of certain sorts
benefiting oneself and others, to the goodness of which actions (desired or not) I shall come in due course.

In cases such as those just described, where one person, A, desires the well-being of a second person, B (e.g. the
satisfaction of some desire of B), and the desire is satisfied, there is a triple good. There is the primitive good for B of B
being benefited, e.g. by the satisfaction of B's desire. There is the greater good for A of the satisfaction of A's desire,
greater because A's desire is focused on something intrinsically good, e.g. the fulfilment of another's desire. And there
is the further good for B that A's desire for B's well-being, e.g. his (B's) desire to be satisfied, was satisfied. We are
fortunate if our well-being gives happiness to others, even if we do not know that it does. We can see this by the fact
that we regard ourselves as fortunate when we discover that another was made happy by our being happy. Although
the discovery causes us to regard ourselves as fortunate, what we regard ourselves as fortunate in respect of is what we
discover (not the fact of our discovering it) — and that is something which could occur without our having

THE GOOD GOALS OF CREATION

discovered it. (Note the similarity of this argument to the argument about beauty in Chapter 3, and other arguments in
later chapters.)

There is special good in the eventual satisfaction of a strong desire which persists through varied experiences and other
desires, even though temporarily frustrated (so long, of course, as it is not for something bad). Hence the special good
of the satisfaction of an animal's desire for the return of its offspring lost for hours or days. The greater good of the
satisfaction of persisting desire is yet greater when what is desired is the success of some action of some kind. It is
good that someone persist in attempting to search for food, and eventually find some; stick at trying to learn to type or
drive or swim and finally succeed.

And note of course that in these as in other cases, the good of the satisfaction of desire does not lie solely in its known
satisfaction. It follows that there is significant good in a persisting desire being satisfied after the desirer's death. The
greater good of the satisfaction of persisting desires is yet greater, the better their object; and so there is especial good
in the satisfaction of shared persisting desires for the satisfaction of the persisting desire of a third individual.

It is not just the satisfaction of desire which is in general a good, and especially when it is of a desire for a state of
affairs good for other reasons. The mere having of desire for anything at all, so long as it is not for a bad state, is good.
It is good for me that I want things, long for things, am inclined to try to bring about things. It is better that I be
someone to whom things matter rather than a ‘cold fish’ who acts under the guidance of reason alone. Desires in
themselves are good, except when they are desires for what is bad; but they are better, the better for other reasons are
the states desired. It is good that I should desire everything that is good; then my system of desire will be rightly
attuned to the values of things. And if my desires are focused on your well-being, that is a good not merely for me but
also for you: how fortunate we are if people care about us, mind what happens to us.

All the features which make good desires better make bad desires worse. It is bad for me and for you that I should
desire your harm — it is bad for me to be a hater; and, as it is good to be loved, so it is bad to be hated — even if you do
not know about it. Joint desires and cooperative desires for states of affairs bad in themselves are the worse for their
joint and cooperative elements. A's desire that B's hatred for C be fulfilled is bad not merely because both A and B

THOUGHT AND FEELING

desire C s harm, but because A desires what is bad for B, the satisfaction of a bad desire. And a persistent desire for the
bad is worse than a casual intermittent one.

The absence of a desire for something good cannot as such be considered a bad state. There is nothing the matter with
me if I do not desire that some child in China of whom I have never heard pass her examination. But the absence of a
desire for a goal which I ought to pursue is, I suggest, a bad state. I ought to educate my children. If I do not mind
whether they are educated or not, if I have no desire for their well-being in this respect, then there is, I suggest,
something bad about me — even if I do my best to see that they are well educated. Obligations are limited and, in
principle, knowable. There is something bad about us if we do not know our obligations; and, if we do know them,
and find it difficult to act on them, then that will be due to bad desires deterring us (see p. 86) and our having these is a
bad thing.

While other-centred desires are clearly a good thing for us — we are the better for wanting other people to be happy — it
is important to note that self-centred desires are also a good thing for the desirer. For, as we have seen, it is good that
you desire my well-being rather than some ethically irrelevant state of me, because the former is a good thing; and in
that case it is good that I too desire it, for it is good that I desire whatever is good. Now it is good that any good desire
be fulfilled. But if my well-being consisted solely in promoting your well-being, and your well-being consisted solely in
promoting my well-being, and so on in a circle, there would be nothing ultimately in which the well-being of either of
us would consist. There would be nothing which I could do to promote your well-being — if doing the latter consisted
in causing you to cause me to cause you to cause me . . . and so ad infinitum. A few really self-centred desires are needed
if there is to be any altruism. If I want to eat chocolate, there is something you can do for me: give me some. And if
you also want some, there is something I can do for you, as well as or perhaps instead of indulging my own desire: give
you some. Only with a bit of self-centredness (perhaps taking a higher form than a desire for chocolate) is there scope
for generosity, for altruism generally, for individuals with differing natures, and for moral choice. But of course self-
centred desires need to be properly balanced, in a morally perfect individual, by strong other-centred desires.

THE GOOD GOALS OF CREATION

I draw attention to one important class of desires and two important particular desires on which I shall need to
comment later. The class is that of instrumentally useful desires: desires to get something which goes with some good,
or to get rid of something which goes with some harm. Let us call the former incentive desires, and the latter deterrent
desires. Biologically useful desires are the obvious examples of instrumentally useful desires. Among biologically useful
incentive desires are desires to eat, drink, sleep, and have sexual intercourse. Such desires are in themselves good to
have and satisfy (except in those cases where the state desired is overall bad, e.g. when the sexual intercourse involves
adultery). But their satisfaction also has further consequences, not necessarily envisaged by the desirer, of biological
utility for individuals and the race; e.g. sleep making the sleeper fresh for the next day, and sexual intercourse leading to
the survival of the race. Among the biologically useful deterrent desires are the desires to avoid pains of various kinds.
When we put a hand into the fire, we get a strong burning pain; this involves a desire that it cease, which can only be
satisfied by removing the hand from the fire. Such desires are again good to have and satisfy (except of course where
the state desired is overall bad, which in this example would only be the case under very peculiar circumstances); but it
is of course bad that this desire should be initially unsatisfied. The satisfaction of the desire has biologically useful
consequences, e.g. that the hand is not destroyed. (But it is not always the case that the sufferer who seeks to stop some
pain realizes the biological consequences of doing so.) Pain arises in the first instance because of its biological utility; it
can also occur when not biologically useful, but it is important to have in mind that that involves its occurrence outside
the context where it has its main use.

The (incentive and deterrent) useful desires may be immediate, or delayed in the sense of being desires for the
satisfaction of some future desire, or not to have some future frustrated desire. When we acquire some knowledge of
how the world works we know that under certain circumstances we shall have desires which we can then satisfy or
ones that we cannot then satisfy. Our present desires to have the former then act as incentives; and our present desires
to avoid the latter act as deterrents. Among the useful delayed incentive desires are the desires that when we desire to
eat or sleep, we shall be able to do so, which lead to our buying food and securing

THOUGHT AND FEELING

accommodation. Among the useful delayed deterrents are the desires that we do not have pains which we cannot there
and then remove: the desire to avoid the hangover we shall get if we drink more than eight pints of beer, influencing us
not to drink those pints now.

There are instrumentally useful desires other than these simple biological ones. In so many (but not all) economic
systems, for example, the desire of one person for fortune is instrumentally useful in developing technology and
bringing prosperity to others. In our world all desires for longer-term goals are instrumentally useful in leading us to
seek and acquire theoretical knowledge. Desire to communicate with distant relatives, to fly in the air, and to cure
disease and prolong life require us to seek knowledge of how the world works which we can then use to achieve our
goals. I have argued that true belief about how the world works, and more so, knowledge in at least one sense of the
latter, is a good thing to have. But while most of us do desire that knowledge for its own sake, that desire is not initially
in many of us a very strong one. Desires for most other good things therefore have an instrumental utility.

The first particular desire to which I draw attention is the very good desire to be thought well of. We like to be liked,
and it is very good that we do. For what we desire — that we shall be valued by others, that they shall seek our company
and respect our opinions and achievements — is a great good for us. And it is a great good for them that we believe
their good opinion worth having, and desire their company. A world in which people could not care less what others
thought of them would be a terrible world. Of course there are occasions on which the achievement of a greater good
(e.g. standing up for justice when that will make us unpopular) may require that we resist the desire to be thought well
of. But to have the desire is good and it is normally good that it be satisfied.

The second particular desire to which I draw attention is involved in many of the good desires to which I have been
drawing attention. It is simply the desire to continue to exist in what is a good state, including happiness. Existing in
what is objectively a good state is a good thing, and even better is enjoying it. So it is very good that we should desire
(in this sense) our own future well-being.

THE GOOD GOALS OF CREATION

Emotions

Many mental states to which we give names are mixed states in involving certain combinations of belief, desire, and
sensation; and those which involve the right combinations are very valuable. The right combinations are ones in which
the desires are to perform actions and have things happen which, given true beliefs, are good to happen, and in which
sensations ‘chime in’ naturally.

An emotion such as anger, for example, includes a belief that someone has done wrong (to oneself or someone else), a
desire to hurt them for that reason, and often also a sensory element of tightening round the eyes and in the stomach,
which we do not welcome, but which ‘fits’ the desire in that satisfying the desire to hurt would involve an ‘uncoiling’
which would get rid of the tightening. There is a ‘righteous anger’ in which we have a true belief that someone has
done a wrong and a good desire to bring this home to them by word or deed (urged on by appropriate sensation),
good for us and (on balance) good for them.

Respect is a good emotion to have towards those who deserve it. It involves a belief that someone is great in some
way — greatly knowledgeable, powerful, wise, good, or has achieved various things (in spite of hardships), for
example — and a desire to acknowledge this. All people deserve to be acknowledged for what they are, and so for the
great-making qualities they possess. I have argued that knowledge (in one or more senses) is a good for the knower,
and I shall be arguing in Chapter 5 that causing, and so power — the ability to cause — is a good for any agent, and
especially for a free intentional agent. Those who have these qualities deserve respect; and for us to fail to show it to
someone would be bad for them and bad for us, for we would not then be rightly attuned to things as they are. Respect
is good if and only if the belief involved in it is true — if its object deserves respect. But if it does, respect is
acknowledgement of worth. Of course if power is misused, the powerful may not overall deserve respect. Yet
someone whose natural inclinations are to be disrespectful of the powerful, cynical about the saints, supercilious about
the learned as such, has got something the matter with them.

Aesthetic contemplation of beauty seems to involve a belief element (the belief that the work of art is beautiful) and a
desire element

THOUGHT AND FEELING

(the desire that the work should exist). True beliefs and right desires correcdy focused on what is in fact beautiful is
clearly a considerable good.

‘Love’ is a word used in many senses. But in many of these senses it involves a whole complex of beliefs, desires, and
sensations, as well as an active component of performing various intentional actions. I confine myself in this chapter to
reflecting on the goodness of the various involuntary components. Loving another involves believing that some
aspects of them are good. It may be that we believe them beautiful, or successful or charismatic, or sympathetic, or
honest, or whatever; but of course it does not involve believing that all aspects of them are good. Centrally it involves
desire: desire that things go well in all ways with the beloved; and desire that we may interact closely with them in
shared activities, and that we may share desires and beliefs with them; desire that we know all about what they desire
and believe, and that we desire and believe the same things. And it involves the desire that the beloved should desire all
of these things with respect to us (that they should desire to share our desires and beliefs etc.). And it involves (though
less centrally) sensations peculiar to the having and satisfaction of the desires: the tinge of excitement at the beloved's
presence, a sensation which we desire to continue; and the sensation involved in frustration at lack of closeness, a
sensation which we desire not to continue. Given that my beliefs about them are true, and that the desires are for good
states, it is good (for them and for me) that I love all animate beings, but especially those with whom I have contact
and in particular those greatly dependent on me or on whom I am greatly dependent. It is good that I love these latter
more than others — otherwise I do not pay proper tribute to their connection with myself; I trivialize our personal
relations. And again, the love that persists despite inadequate satisfaction has its own special greatness. And if the
beloved suffers, it is good that the lover feels frustrated in his own desires. And if the beloved dies, it is good that the
lover mourn the beloved. We could be so made that immediately we lost a loved one, we ceased to miss him or her.
But how horrible that would be! If love is good, it matters that we grieve, that we pine for the loss of a loved one. It
constitutes the proper tribute of a lover to the lost loved one. Someone is indeed unfortunate if no one is sorry when
they die. A love is only worth having if it is still there when unsatisfied; that is what constitutes its

THE GOOD GOALS OF CREATION

serious nature. What goes for grief goes also for so many other unsatisfied longings: if what is longed for is good, then
there is good in our caring that we do not have it. We creatures of desire pay with our desires the tribute of our nature.

It is good that I feel joy at the other's success, at being close to them; and that I feel compassion for the sufferer,
sadness at failure, and grief at the loss of the departed (both in this package which constitutes love, and more
generally). Again, these attitudes are both good for him who has these right attitudes and good for him on whom they
are focused — very good if the latter knows about them and that gives him some pleasure, but good even if he does
not — for the reason given earlier. What gives him pleasure is the knowledge of the existence of a good state (e.g.
someone else's compassion for him), and that state can exist without his knowing about it.

The love of embodied persons, such as humans, for each other will naturally involve a bodily element — for bodily
proximity (so that we may desire sensations from the other's body, being that material object over which the beloved
has most immediate control; and may convey sensations in the same way to the beloved). But sexual desire — to derive
sensations out of a very special kind of interaction with the other — is of course not necessarily involved in love as such;
and, notoriously, may exist without the other aspects of love. But of course it can and often does go together with
other aspects of love in a wonderful package.

Human and other finite beings can only interact with a limited number of other beings, and while it is good that they
shall love many others, only a love for a few others can possibly be satisfied. But, quite apart from that limitation of
finitude, there is special value in special kinds of love which are not shared. It is a very great good for someone A to be
loved by another B in a way and for qualities for which B loves no one else. If we are valued only for qualities which
can be shared with others, our particularity — what makes me me — is not recognized. The qualities which cannot be
shared are those which consist in interactions at particular times with particular individuals. If I am valued for being
someone's first-bom child, or spouse, or sole doctoral supervisor, then I am valued for something special to me.

Only emotions which are rightly focused are good. An anger driven by a false belief that wrong has been done, or in
which the

THOUGHT AND FEELING

desire to hurt is quite out of proportion to the wrong done, is clearly bad. So too is respect for those who do not
deserve it. Love too can sometimes be misplaced. A love which (is not merely caused by, but) has as an essential
constitutive part a false belief is not good: if I love you because I believe that you are my faithful spouse but in fact you
are cheating on me badly, then my love is not good, possibly even bad. And if I love you for something which is bad in
you — if I love you for your cruelty — that is a bad love. And a love which if realized would spoil another love — such as
a love which seeks a sexual relationship with the wife of someone else to whom she is happily married — is also clearly
bad. But of course in general love is good — very good indeed, the best of all the emotions.

True beliefs, good desires, and thereby rightly focused emotions are good for the individual who has them,
independently of their consequences; and good for the individuals on whom they are focused. It is good that they
should be interdependent in the various ways described, and very good that they should be fulfilled. This has been the
argument of this chapter. Hence one would expect a good God to create individuals who have such beliefs and desires;
and (if there is a God) clearly he has done so in abundance.

The higher animals have all of the simpler kinds of belief and desire which I have described, and humans have the
more sophisticated kinds of belief and desire. One might come to think that, in that case, because the beliefs and
desires which humans can have are better than those which animals can have, God would create only humans. But that
thought would be mistaken for more than one reason. It will suffice to mention here just one reason; I will come to
others in the next chapter. This one reason is that among the great goods available to humans are those of responding
in the right way to those with lesser abilities and so to love and to care for animals. Although animals can respond with
great love to human love, they cannot love in the same way; and hence human love of animals will have a uniquely
valuable character (because reciprocable only in a limited kind of way). To be able to love those who cannot love you in
the same way is to have a unique opportunity for generous love. ‘If you love them that love you, what reward have
you? Do not even the publicans the same ?’ 76

76 Matt. 5: 46. The context of verses 43—8 suggests that the ‘reward’ is not an after-death reward, but the present ‘reward’ of being perfect On the general ■
interpret the ‘rewards’ promised by the Sermon on the Mount, see Ch. 13 .

: of how to

THE GOOD GOALS OF CREATION

There are, however, many false beliefs and bad and frustrated desires of the types which I have described, both among
animals and — more strikingly — among humans. The pain of incurable disease and accident, grief and disappointment
at failure, the feelings of hatred which individuals and nations have for each other, are all too evident to need much
further comment; and we shall need to inquire in due course why God should allow these to occur.

In general the Christian tradition has recognized and praised God for all the true beliefs and good desires of the types
described. But at times there have been thinkers and groups of Christians who have judged that desires focused on
mere human goods (as opposed to desires for closeness to God) are not good. 77 Of course all the tradition has
recognized what I have stressed earlier, that the desires for mere human goods need to be kept under control — we
must not yield to desires for food or sexual intercourse under many circumstances (e.g. when others need the food
more, or when the one sexually desired is married to another) — and that desires for what it is bad that we should have
are themselves bad. And because it has recognized that it is often difficult for humans to get their desires rightly
attuned to the good, it is often necessary for them deliberately to frustrate desires which tend to exert in general more
influence than they should, even when it would otherwise be good to indulge them. Thus the tradition has commended
fasting in Lent, abstaining from food which it would otherwise be good that we should have, in order to be able to
keep our desire for food under control during the rest of the year, so that it does not influence our conduct more than
it should. But the tradition has on the whole maintained that desire for mere human goods and their indulgence is as
such a good thing. The Christian thinker who above all said this very clearly was St Thomas Aquinas. Question 2 of
part 2(1) of the Summa Theologiae discusses what human blessedness ( beatitudo , often misleadingly translated ‘happiness’)
ultimately consists in. Aquinas's final answer is that it consists in the Vision of God. But, before reaching that answer,
he considers whether it consists in riches, honours, fame, power, bodily well-being, sensory pleasure, or in ‘some good
quality of the soul’ (e.g. intellectual achievement).

This tendency in Christian thought reached its culmination in the writings of some later medieval mystics and in the Quietist movement of the 17th century, which denied
the goodness of any desire, even desire for God. On this tendency, see K. E. Kirk, The Vision of God (Longmans, Green, 1931), 451-66 .

THOUGHT AND FEELING

With respect to each of them, he holds that they are goods, but they are lesser goods: blessedness cannot consist
‘principally’ in being honoured, the ‘perfection’ of human good cannot consist in being acknowledged by other people,
and so on. But all these things, even ‘external goods’, are ‘required’ for the ‘imperfect blessedness open to us in this
life ’. 78 And since all these things are good for us, it is good that we should desire these things.

5 Action

And God said, Let us make man in our image, after our likeness: and let them have dominion over the fish of
the sea, and over the fowl of the air, and over the cattle, and over all the earth, and over every creeping thing that
creepeth upon the earth. And God created man in his own image, in the image of God created he him; male and
female created he them. And God blessed them: and God said unto them, Be fruitful, and multiply, and
replenish the earth, and subdue it; and have dominion over the fish of the sea, and over the fowl of the air, and
over every living thing that moveth upon the earth. And God said, Behold, I have given you every herb yielding
seed, which is upon the face of all the earth, and every tree, in the which is the fruit of a tree yielding seed; to you
it shall be for meat: and to every beast of the earth, and to every fowl of the air, and to every thing that creepeth
upon the earth, wherein there is life, I have given every green herb for meat: and it was so. And God saw every
thing that he had made, and, behold, it was very good. And there was evening and there was morning,
the sixth day.

(Genesis 1: 26-31)

Causing Spontaneously and Causing Freely

In the last two chapters I was concerned with the goodness of involuntary states — both non-conscious and conscious.
Now I move on to consider the goodness of action, that is causing or bringing about. All causing is, as such, good (for
the agent); even if it is unintentional (of plants say, which produce seeds to produce new plants), but especially if it is
intentional. An intentional action is one which an agent does because he or she means to do it. Causing

ACTION

is as such a good thing because in causing the agent contributes. It is only causing what for other reasons is bad, which
is itself bad. Causing what is good, e.g. the states of affairs described in the last two chapters, is very good; and not just
for the reason that those states of affairs are good, as examples will bring out in due course.

An intentional action may be objectively or subjectively good, or both. It is — to repeat from Chapter 1 — objectively
good if it is good in its nature or (likely) consequences, apart from what the agent believes about it; and typically this
will be a matter of its consequences. Other things being equal, an action is good in virtue of having good effects. It is
subjectively good if the agent believes that in the respect in which it is intentional (i.e. in respect of what he or she is
seeking to achieve) it is a good action; and is doing it for that reason. 79 An intentional action is the better for being both
objectively and subjectively good.

An objectively good intentional act is good, even if it is not done intentionally in the respect in which it is good.
Performing some intentional action (e.g. I buy your house), I may unintentionally do something else good (e.g. I save
you from bankruptcy). Such a good action will derive its goodness not from its intention but from its effect. What is
achieved is a good, and it is also a good for the agent who effected it, even if unintentionally; he is lucky to be a vehicle
of benefit. His agency is not a wasted agency; it contributes.

Conversely, an unsuccessful action which aimed at something believed good is also a good for the agent. It is good that
people try to help the starving even if they do not succeed. We can perhaps see the good of unintended success better
than in my earlier example by bringing in this latter point, and considering an action which is aimed at a good but fails
and is better for having an unintended good result than it would be otherwise. I toil to save a life and fail; but the
record of my efforts makes possible the development of a technique for saving other lives in future. Is that better than
if I failed and by chance someone else hit on the new technique? Yes, because my efforts are crowned, and therefore it
is a good for me, not merely for those whose lives were ultimately saved. But clearly things are better if the good which
I achieve is intended. While it is good that I try to feed the starving even if I do not succeed, it is better if I do
succeed — but not just because it is good that the

For fuller analysis of the distinctions of this paragraph, see my Responsibility and Atonement (Clarendon Press, 1989), ch. 1 .

THE GOOD GOALS OF CREATION

starving have enough to eat; it is a good for me that I am privileged to help them, that I am of use. It is even good that
I aim at something believed good, even if falsely — for I am better for a direction of will towards the good, even if
somewhat misinformed about wherein it consists. 80 But it is better if my moral beliefs are true; and, while keeping in
mind the possibility of false moral beliefs, I shall for the present, for simplicity of exposition, assume that agents have
true moral beliefs; that is, true beliefs about the moral worth (good or bad) about the consequences (intended or
foreseen) of their actions.

An intended good action, when fully caused, is the better if the agent is fully behind it; if the inclinations (i.e. desires)
which lead him to act are not in conflict with each other. We value the spontaneous pursuit of the good, the pursuit of
the good which the agent fully desires to pursue. We value the willingly generous action, the naturally honest,
spontaneously loving action. But we value even more that the pursuit of the good should result from a free choice of
the agent between equally good actions, that is, one resulting from the exercise of (libertarian) free will. It is a good for
any agent to have such a free choice; for that makes him an ultimate source of the way things happen in the Universe.
He is no longer totally at the mercy of forces from without, but is himself an autonomous mini-creator. And in both of
these cases — the spontaneous good action and the freely chosen good action — the action is better if efficacious, if it
does make the intended difference to how things are in the world outside the agent.

But the good of freely forwarding the good is better if the agent has a free choice between a greater and a lesser good;
better still if he has a free choice between the good and the bad, and even more so if the possibility of doing bad
includes the possibility of doing wrong (see Chapter 1 for this distinction), not just between alternative equal goods. (In
this discussion I am counting ‘doing nothing’ as one alternative open to an agent.) I shall call free will involving a
choice between good and bad ‘serious free will’, and if wrong is also a possibility, ‘very serious free will’, to be
contrasted with ‘unserious free will’, where the choice is only between alternative

For reasons which I give in Responsibility and Atonement , ch. 1, 1 do not think that anyone can be totally mistaken in their beliefs about which actions are good. To have a
concept of the good, you need quite a few true beliefs about which actions are good.

ACTION

goods (and if the goods are equal goods, Very unserious free will’). An agent who has serious and efficacious free will
is in a much fuller way an ultimate source of the direction of things in the world; he is not pre-programmed to make a
choice within a range of choices all of the same moral kind — good. And he is an ultimate source in an even fuller way
if the choices open to him cover the whole moral range, from the very good to the very wrong. Having the opportunity
thus to influence the way things go is a great good for the agent, whether or not he exercises it in favour of the good.
But an agent who exercises his very serious free will in favour of the good is blessed indeed. He rejects contrary desire
and directs the flow of events in favour of the good rather than the bad. The agent has used his choice to make a
significant difference to the world for good.

Now agents, such as ourselves, capable of recognizing reasons for doing an action and so something good about doing
the action, will always do the action which they see the most reason to do, i.e. regard as the best to do, unless some
influence diverts them from that. To have the belief that there is a reason for doing an action is to acknowledge that,
thus far, it would be sensible, appropriate, reasonable, rational, to do the action, that it is the thing to do. Really to
believe that some action would be sensible, appropriate, etc. to do is to acknowledge, to put the point dramatically, the
summons of the action to me to do it; and thereby to have an inclination to do the action, other things being equal. I
could not recognize R as a reason for doing A unless I accepted pressure from how I see things to be in the direction
of doing A. To admit that R was a ‘reason’ for doing A but to deny that I had any inclination at all in consequence to
do A would be to say something apparendy contradictory, and to suggest that, when I said that R was a ‘reason’, I was
using the word in an ‘inverted comma sense’ and meant something like ‘what most people would consider a “reason” ’.
To recognize a reason for doing A is only to have an inclination to do A, other things being equal. But other things
may not be equal. There may be other and better reasons for not doing the action. To believe that there is reason
overall for doing the action is to believe that, on the balance of reason, it would be sensible, appropriate, reasonable,
rational to do the action. I may still not do the action because I may yield to non-rational forces which influence the
purposes I form. But to believe that the balance of reason supports doing it entails

THE GOOD GOALS OF CREATION

being inclined to do it, and doing it in so far as one is unimpeded by non-rational forces . 81

It follows that if an agent is to have the option of doing what he regards as less good or bad actions — and so, on the
assumption that he has true moral beliefs, doing less good or bad actions — he must be subject to a stronger desire to
do an alternative action. (The alternative may of course simply be the action of ‘doing nothing’.) With such desires, i.e.
temptations, alone can he have a free choice of pursuing the best or not pursuing the best. We regard as blessed and
honour those who freely do the good when they overcome the temptation to do instead the bad or less good.

Just as if reasons alone influence action, an agent inevitably does what he believes to be the best, so if desires alone
influence action, an agent will inevitably follow his strongest desire. Free choice of action therefore arises only in two
situations. One is where there is a choice between two actions which the agent regards as equal best and which the
agent desires to do equally; which — on the assumption that his moral beliefs are true — is the situation of very
unserious free will. The other is where there is a choice between two actions, one of which the agent desires to do more
and the other of which he believes it better to do . 82 This latter is the situation of temptation. Temptation may come in
three forms: where the desire to do an action other than that believed best and which is stronger than the latter is a
desire to do a less good action, or a bad but not wrong action, or a wrong action.

For any situation of temptation there are possible different degrees of free will. Our desire to do the bad (or whatever)
may be only marginally stronger than the desire to do the good and in that case it does not require great effort of will
to conform to the good. On the other hand the temptation, e.g. the desire to do bad, may be almost irresistible;
immense effort will then be required if we are to conform to the good.

31 See Additional Note 3 .

32 There is a further circumstance under which free choice is possible. That is where an agent has a choice between an infinite number of good actions, each of which is, he
believes, less good than some other such action; there is, he believes, no best or equal-best action. However, since only a being whose power is unlimited in some respect will
be in such a situation (as, I elucidated in Ch. 1 , is God), humans or animals on Earth are never in such a situation; and so for the sake of simplicity of exposition, I ignore
this possibility in the text at this point.

ACTION

How is the value of the different forms of free will (serious, unserious, etc.), and different degrees of each with respect
to a given action, to be weighted against each other? Fairly clearly to do good out of very serious free will despite
strong contrary temptation is the best exercise of choice. It is a marvellous thing for an agent to be the ultimate source
of things going well rather than very badly. And the agent is more ultimately the source of that happening the more the
passive forces at work in the world were sweeping it along in the wrong direction and by his effort alone the current
was turned in a different direction. To generalize: the more serious the free will and the stronger the contrary
temptation, the better it is when the good action is done. So much seems fairly evident. But the agent's having good
desires is a good in itself, and so when they are added to the good of the good action which follows naturally from
them, it becomes unclear just how much better it is overall to do a good action out of serious free will rather than
generously and naturally. Fairly evidently too if the less good, bad, or wrong is chosen freely, the free exercise of choice
is worse, the worse the action chosen (e.g. if it is wrong rather than merely bad), and the less strong the temptation to
do it. (Then the agent has less excuse.) And the desires for the less good, the bad, or wrong which make possible the
choice of the latter are bad. But also, as I have already suggested, the very fact of the agent having a free choice is a
great good for the agent; and a greater good the more serious the kind of free will, even if it is incorrectly used. It is
good for the agent that the world goes the way he chooses rather than the way someone else imposes on him.

All the above, I suggest, most people can accept. But where views will differ in a way not easy to resolve is in how
these different goods and bads are to be weighed against each other. To give someone free will makes it up to him
what will happen. That is a great good, and (from the viewpoint of his creator, who does not make his choice for him)
there is a certain chance of a further good that he will use it to choose the best. Both of these — the actual good of the
free will and the chance that it will be used for the best — have to be weighed against the chance that the agent will do
the less good, bad, or wrong act, and the badness of the bad desires which make the latter choices possible. The chance
that the agent will do (what he believes to be) the best will be greater, the less is the temptation to do the alternative.
But the less the

THE GOOD GOALS OF CREATION

latter temptation, the less is the goodness of the act done in resisting it. I have no easy algorithm for working out which
kind of free will is the best to have.

The good of the good free act, and the bad of the bad free act, will be greater if the act is efficacious than if it is not (i.e.
if we succeed in doing the act rather than merely attempt it but fail), both because of the goodness or badness of the
effect, and because the agent is the instrument of it. The extent of the possible consequences for good or ill will clearly
affect the equation, make a difference to whether it is a good or bad thing for a creator to give creatures free will of a
certain kind. But in itself (i.e. apart from the bad desires which are needed to make it possible, and the chance of bad
consequences) I suggest that having a free will which can make a big difference to the world is a greater good than
having one which can only make a small difference. In a later chapter I will have to face the issue of how great a chance
of how much bad is too much for God to give us a certain sort of efficacious free will. But meanwhile I wish to reflect
on the goodness of various good actions which it is good that agents should do spontaneously, and better that they
should have the opportunity to do freely rather than various alternative actions. The range and importance of the
effects which they can bring about constitutes the significance of their actions. If those actions result from free choice,
then the choices have corresponding significance. I shall talk of agents having responsibility for things in so far as the
way those things go results from their free choice. Agents have great responsibility in so far as they can make free
choices of great significance. ‘Responsibility’ simpliciter . ; as I shall use the term, belongs only to free agents, though I may
need to talk of unfree agents having an unfree responsibility for others when their actions have significance for those
others (as do the actions of animals for their offspring). All I am claiming at present is that it is good in itself (i.e. good
for those agents) that free agents have such responsibility, even if the chance of their misusing it is so great and the
resulting bad so great that on balance it is not a good thing that God give agents such free will.

So in summary the goodness of having free will is a function of its freedom, namely its kind (serious, unserious, etc.)
and degree (strength of temptation); and its significance (the greater the importance and range of difference its exercise
makes to the world; the difference made by it being efficacious). The significance of a

ACTION

choice is a matter of the amount of responsibility for ourselves and others involved in it. Given that we have true
factual beliefs about the consequences of our choice and true moral beliefs about their moral status, any serious choice
will be to some extent a significant choice. For it will mean that our choices are not merely good or bad in themselves,
but make a further difference in the world for good or bad. But choices between good and bad may be choices which
have big (good or bad) effects or small ones. The importance and range of difference we can make to the world
constitutes the significance of our choice. The goodness or badness of the exercise of free will, as opposed to merely
having it, will of course depend also on the goodness or badness of the choices actually made and the difference which
they make to the world.

I assume throughout most of this chapter that agents have true moral beliefs. The goodness of an agent's choice is
clearly largely a matter of its subjective goodness, its choice between the alternatives as he views them; what is good is
that the agent shall make the choice which he believes to be the best. But, as the discussion on pp. 83-4 showed, the
goodness of the efficacious free act depends, as well as on the goodness of the choice, on the objective goodness (not
believed goodness) of the act — and typically this will be a matter of the goodness of its actual consequences (not just,
to repeat, the goodness or badness of those consequences for those to whom they happen, but their goodness or
badness for the agent as the agent of those consequences). In cases where the agent has false moral beliefs, we must
make the corresponding adjustment to the equation, taking into account also the fact (considered in Chapter 4) that
false moral beliefs are in themselves (and so for the agent) a bad thing.

The Scope of Responsibility

It is good that agents have the opportunity to make the world beautiful or ugly; we reflected briefly in Chapter 3 on the
goodness of the beautiful states. It is good that agents have the opportunity to benefit or harm themselves, in the short
and long terms, by imposing on themselves any of the good or bad states described in Chapter 4. These opportunities
will include the primitive giving of pleasure in the short term (e.g. by eating chocolate) or the primitive allowing

THE GOOD GOALS OF CREATION

of pain (by not bothering to remove a stone from one's shoe); and of taking steps to provide for long-term happiness
or through negligence allowing long-term unhappiness (e.g. by not taking steps to cure an incipient disease). But, as we
saw, there is more to the good than what is enjoyed; and there is more to the bad than what is disliked. And it is also
good that animals, albeit unfreely, should benefit themselves, each other, and humans in as many of these ways as their
nature allows.

It is good that the choices open to agents should include the choice of whether or not to seek to acquire knowledge (in
the objective internalist sense) through observation, experiment, and reflection; rather than that they should be born
with all the knowledge they can ever have or acquire it by accidental observation or being told by some expert.
Knowledge of empirical and moral matters, I argued in the last chapter, is valuable for practical purposes. I may
initially have no idea what will be the effects of some of my actions, short- or long-term; and among the choices which
it is good for God to give me is whether to bother to find out or not to take the trouble. I have the choice of
investigating, by seeing the effects on other people and asking experts, the effects of different kinds of diet and lifestyle;
or of not bothering. It is good too that I should have the opportunity of working out for myself whether some of the
effects my actions have are good or bad, rather than have true moral views foisted upon me. In a liberal society we
have come to see that it is a good that people should make up their own minds about moral issues. That need not
imply that there is no truth about what is morally good or bad — only that it is good that we have the opportunity to
choose freely to seek to discover those moral truths for ourselves, or, alternatively, not to bother; and to admit them
honesdy to ourselves when we find them, or to try to hide them from ourselves . 83 Moral inquiry takes the form of
talking to other people of different kinds, hearing about their experiences, becoming involved in their lives; seeing
television programmes, reading novels, and even reading books of moral philosophy.

While it is unavoidable which beliefs we have at a given time, we may refuse to admit some of them even to ourselves (see The Evolution of the Soul. \ rev. edn. (Clarendon Press,
1997), 292—6 ). And it is always up to us whether we choose to investigate further the truth of some proposition, and so open ourselves to new evidence which will lead us
(by involuntary processes) to change our beliefs; but if we investigate, we cannot predetermine the results of our investigation.

ACTION

Animals cannot recognize moral truth, but the higher animals can discover empirical truths, and it is good that they
should do so; and it is good that they should learn not merely by accidental observation or being taught, but that they
should actively seek to find out for themselves where the food is, what food is nourishing, and so on, rather than just
being born with that knowledge. I claimed in the last chapter that it is intrinsically good for us — not merely
instrumentally good — to have both moral knowledge and general theoretical knowledge, and that is a further reason
why it is good that we should have the choice of seeking such knowledge or not bothering to do so. Theoretical inquiry
into fundamental matters of no immediate practical importance — such as the origins of society and the Universe, and
the forces at work in it — and the deepest philosophical thinking about it involves cooperative study and discussion
over centuries.

It is good too for agents to have the choice of whether to increase their power or not to bother. In order to have this
choice we need to know how to increase our power; and getting into that situation may involve inquiry. The acquisition
of knowledge may by itself give us power: knowing how a radio works may give us the power to build one. But it may
be necessary to do other things as well: build our muscles to increase our bodily power; practise the language or the
musical instrument in order to be able to speak or play it well.

A person's character is her system of desires and beliefs (principally moral beliefs); and just as it is good that agents
have the choice of seeking to improve their beliefs or of not bothering to do so, so it is good that they should have the
power to modify their desires over time. It is good that they be able to develop good desires and strengthen the best
desires (including, for example, the desire for knowledge); or allow themselves to be captured by bad desires. But,
given my earlier analysis that a free choice between the (believed) best action and lesser goods and bad and wrong
alternatives involves desires for the latter stronger than desires for the former, such character modification will involve
a restriction of the kind and degree of free choice. In so far as we eliminate bad desires, our freedom of choice will be
restricted to good alternatives.

One way in which this could happen is the way it does happen in humans. Humans are so made that, by forcing
themselves to do good actions when it is difficult, it becomes easier and easier to do

THE GOOD GOALS OF CREATION

them, until finally we desire to do them — our inclinations naturally lead us to do them. As Aristode famously
remarked, ‘we become just by doing just acts, temperate by doing temperate acts, brave by doing brave acts’. 84 We can
so dedicate ourselves to doing good by constant commitment over time that bad desires cease to have any influence
over us.

And it does seem to be also the case that we can yield so frequendy to whatever desire happens to be the strongest that
we lose the habit of choosing between desires on the grounds of the goodness of acting on them. So often our
strongest desires are bad desires; and there is good in us beginning life like that, for only so, as we have seen, do we
have a choice between good and bad. But if we allow ourselves continually to yield to those desires, we close the
possibility of choosing the good because it is good. We cease to have a free choice between alternatives on the basis of
their overall goodness (that is, on moral grounds). We become a theatre of competing desires, in which the strongest
(and so often the worst) wins. It does seem to be the case that the evil men of history have got themselves in this
situation. I shall need to say a little more about the condition of such persons in Chapter 8, to consider whether it
involves the elimination of conscience. Meanwhile, I note the immense significance of the choice of character
formation open to us. The choice of our own characters involves many very serious choices over time and is in itself
clearly of the deepest significance, quite apart from the consequences for others which may follow from our good or
bad characters.

One of the greatest goods which agents can have is responsibility for others. We recognize this when we recognize it as
a good gift to our own children to give them the responsibility for things, animals, and even other humans, and do not
pressurize them too much as to how they are to act. The paradigm situation of responsibility for others is the situation
of the parent. We humans have the awesome responsibility of choosing to bring into the world another human
individual — initially helpless, without knowledge and without free choice — and of helping that individual to understand
the world, to know what is good and bad, perhaps also to develop a capacity for very serious free choice; but also to
help that individual to resist bad desires and to encourage good ones; so that in turn he or she can help yet others to do
the same.

Nicomachean Ethics 1103 d .

ACTION

So just as (for limited periods) a human parent may entrust a younger child to the care of an elder child (despite the
risk that the elder child may abuse that trust and hurt the younger child) because of the goodness of the elder child
having responsibility, so God may entrust creatures to the care of each other. It is good that God be generous in giving
creatures such as us a share in moulding the world and each other, and the deep responsibility that that involves, but if
we are to have such a share, good and bad must both be available to us. Otherwise God would not have given us a
choice of great significance; he would have reserved for himself the all-important choice of the kind of world it was to
be, while simply allowing us the minor choice of filling in the details. He would be like a father asking his elder son to
look after the younger son, and adding that he would be watching the elder son's every move and would intervene the
moment the elder son did a thing wrong. The elder son might justly retort that, while he would be happy to share his
father's work, he could only really do so if he was left to make his own judgements about what to do within a
significant range of the options available to the father. A good God, like a good father, will delegate responsibility.

The responsibility of humans for their children is indeed an awesome one. But it is good that the responsibility of free
creatures for each other should not be limited to responsibility for children. It is good that there be a web of
interactions of different kinds and degrees between creatures, in which they have responsibilities of various kinds and
degree for each other. And of course it is like that with the human race. Children eventually become responsible for
their aged parents. We have limited responsibilities for other relatives, and some of us can often make big differences
to the societies in which we live, through teaching, community work, or political campaigning. Our influence is in no
way limited to the present. The way we educate our children makes a very great difference indeed to the way in which
they educate their children. And so on.

When A benefits B, it is of course a primitive good for B. If A educates B, the acquisition of knowledge is a good for B.
But, I have also been claiming, it is a good for A that he is privileged to make all the difference to B. I need now to add
that it is a further good for B not merely that B acquires knowledge, but that someone else is the intentional vehicle of
that knowledge — has taken an interest in him. We are fortunate if people mind about us, and the

100

THE GOOD GOALS OF CREATION

natural expression of minding is seeking well-being. Sometimes those who ‘don't like to be beholden to others’ do not
see this. ‘I wish that I were not so dependent on my parents for money 5 , says the undergraduate. But ‘so’ is the crucial
word: dependence can come in irksome forms or be too complete. What the undergraduate understandably resents is
his own impotence to benefit himself, not that the good comes to him from a parent rather than from some chance
lottery. And he may also rightly resent it if the parents give unwillingly or stingily. But how awful it would be for us if
nobody ever cared for us enough to give us anything. Fortunately, if there is a God, no human or animal is ever in that
situation.

Just as it is good that there be a criss-crossing of desires and their satisfaction, so it is good that our actions support
each other's actions, and cooperate with each other's actions in forwarding further goods. It is good that my wife and I
pursue a common good — the happiness of our daughter. It is good that we pursue it, not independently, but each
contributing to a total direction: she does this for her and I do that; or, better still, I do this which is no use on its own,
and she does that which is no use on its own, but together the two bits make our daughter happy. It is good for me, for
my wife, and for our daughter.

Some of the good actions which I have been describing are obligatory, others are supererogatory. Obligations, we
noted in Chapter 1, arise from voluntary undertakings (explicit promises, or implicit undertakings — as when in
begetting children, I become obliged to feed and educate them) or from benefits received (as is the obligation to care
for parents who have cared for us).

One great good which an agent can do is (given that the promise is not to do something bad) to make an explicit
promise. He has the power thereby to make some future action obligatory which would not otherwise have been
obligatory; and some future action wrong which would not otherwise have been wrong. If I promise to pay you £100
next week, I make it obligatory that I pay you and so wrong if I do not. Having this power over the moral status of
future actions is a good for creatures; and exercising it constitutes a gift which one gives to another. If my promise is to
you (as opposed to a mutual friend) that I will pay you £100, you now have it in your power to hold me to the promise
or let me off it (i.e. keep it obligatory, or make it no longer obligatory that I pay the money).

ACTION

101

Promises are promises to others, from the obligation to keep which they can dispense us. But there is another form of
commitment which I call Vows’; the vow-taker gives a commitment in a form which gives no one the right to dispense
him from it. By making the vow to do action A, either absolutely or under certain circumstances (e.g. if another person
does action X), the vow- taker makes it a bad act not to do A (at all, or under the relevant circumstances). If not doing
A would be a bad thing anyway, the vow makes it bad for a further reason. 85 There are two reasons why anyone should
make a vow. The first, which only applies if not doing the action would be bad for other reasons, is to make it harder
for the vow-taker to avoid doing the action when or if the relevant time comes. Suspecting that I shall not take exercise
otherwise, I make a ‘New Year resolution’ to go for a run once a week. The second reason is that by making the vow
known to others, that serves to influence their behaviour. If the state solemnly vows (by passing a law which cannot be
repealed and whose application does not depend on the discretion of judges) that all who commit a certain crime will
be punished in a certain way, that will deter people from committing that crime. But it will only do so if people believe
that judges will execute that penalty, even if those judges believe that that penalty is not the best one (for the particular
criminal or for the future of society). Deterrence only works if people believe that the penalty will be executed whether
or not other later reasons suggest that it is a good thing that it should be. One way in which people will come to believe
this is by seeing that that is what has happened in the past. But with a new authority or new circumstances or a new
kind of penalty, we cannot infer from what has happened in the past that a particular penalty will be imposed. We can,
however, infer it in so far as the authority has vowed to impose it, and we have reason to believe that the authority is
good and regards vow-keeping as good.

A promise to do something bad is not, however, a good thing; though if we have the right to give it, giving it surely
does make it obligatory to fulfil it. It may not be good to give you £100 if you will only waste it; but if I promise to give
it, I ought to give it. But a

85 Failing to fulfil a vow can only be a bad thing, and not a wrong thing, if I :
of this point, and the role of vows, see Responsibility and Atonement, passi

right m my contentiot
and esp. pp. 104-6.

l is always obligation to someone else

s. For fuller discussic

102

THE GOOD GOALS OF CREATION

promise to do what we have no right to do cannot bind. A promise to murder your mother-in-law is not merely bad to
make but can create no obligation — such is the surely correct traditional view. Likewise a vow to do something bad
would be bad to make; but, once made, it is good to fulfil it unless we have no right to do the bad action. If that were
not so, vows by morally good authorities to punish for crimes which could not be repeated would have no deterrent
force.

For consider the case of the utilitarian parent who wishes to deter his child from doing a certain wrong act of a non-
repeatable kind. The child has stolen from the parent in the past and the parent has left only one valuable item — a
picture, say, worth £1,000 — and no prospect of acquiring any more valuables nor any income apart from that needed
for immediate expenses. The parent wishes to deter the child from stealing the picture, selling it, and spending the
money on drink and drugs. So the parent threatens, Tf you steal the picture I will punish you’. But the child knows that
the parent is a utilitarian and that if the child does steal the picture the parent will then consider whether he should
punish or not according to whether there will be subsequent beneficial effects of punishing, e.g. whether punishment
will deter from future crime. But if the child has stolen the parent's only valuable item, sold it, and wasted the proceeds,
there will be no possibility of future crime from which to deter. It is only if the parent believes that the threat
constitutes a vow which it would be good to keep despite the badness otherwise (i.e. on pure utilitarian considerations)
of imposing the punishment, and the child believes that the parent believes this, that deterrence will work. And a good
parent could only make such a vow if it is good that vows should be kept when it would be for other reasons bad to
keep them. But of course a vow to do something which the vow-taker has no right to do does not make it obligatory,
or even good to do it.

Vows, like promises, are then yet more ways in which agents can influence the future. Vows and promises are not
always made by uttering such formal words as T vow 5 or T promise’; but may be made in any way which shows that
the agent has committed himself in the ways described here. I shall need to return later to the particular way in which a
vow can influence the future.

There are certain sorts of good action which can only be done in the face of bad states of affairs. Showing sympathy to
the suffering

ACTION

103

(which I distinguish from being in the passive state of feeling compassion), helping the suffering, and showing courage
of a certain sort are like this. I cannot show sympathy to the suffering, nor help the suffering unless there is suffering,
nor bravely bear my pain unless I have pain to bear.

Other good actions which can only be responses to bad states of affairs include those which are responses to the
malevolent or negligent actions of others or ourselves. I can only successfully apologize or make reparation to you if I
have harmed you; or forgive you if you have harmed me and apologized. Other good actions are the appropriate
responses to good states of affairs. There is an obligation to show gratitude to those from whom we have accepted
benefit, and also to acknowledge the achievements of those whom we have benefited and who have put those benefits
to good use. 86 1 claimed earlier that it is a good for me if my efforts to achieve some good are successful. I am indeed
fortunate if my pupils all pass their exams. They have done me a good turn, and I ought to acknowledge that. So more
generally communities should recognize and honour those of its citizens who have put to abundant good use the
nurture and education which it provided for them.

There are also actions similar to these just described which are not obligatory but supererogatorily good. It is good to
show gratitude beyond the minimum obligatory; good for a wrongdoer to give reparation and penance beyond the
minimum; and good to show respect in recognition of the worth and achievement of others, even when the
achievement has no connection with ourselves. So many of these actions are actions which acknowledge the past, in
which an agent acknowledges his identity with a certain past agent and the identity of others with certain other past
agents. They consecrate past actions by giving them a present significance; and so they sanctify history. History does
not get lost, it matters. Wrongdoing is not left as a blot, but wiped out in so far as it can be. Gratitude too consecrates
the past; and the making of promises also makes for bonds of significance crossing the passage of time, which are
strengthened as the promises are fulfilled.

If responsibility is a good thing, then (barring the chances of bad consequences, to which we shall come later) the more
of it the better. It is good that I be able to influence not only my family now,

notions as reparation, forgiveness, gratitude, an

For analysis of such

id reward, see "Responsibility and Atonement , chs. 4 and 5.

104

THE GOOD GOALS OF CREATION

but, through cooperative action, distant communities and future generations. Among the choices we have, both as
individuals and (in cooperation) as a community, is, as I have already noted, whether to seek to grow in knowledge. On
the large scale, this means that we have the choice of whether to invest money in universities and research institutes, to
find out about our past history, the laws of nature, and the truths of philosophy; or whether not to bother. 87 We have
also the choice of seeking (partly through growing in knowledge) to grow in power. We were able to choose whether to
take steps to learn how to ride horses, build boats, make combustion engines and computers — and so to go places,
build things, and destroy things. And by our choices now — to pursue certain kinds of scientific research and invest our
wealth in certain sorts of technology — we may be able in centuries to come to affect for how long the human race lives
and on which planet. To have the choice of whether or not to seek to grow in knowledge and power, we must start
from a position of relative ignorance and impotence; and there has to be a way which we can recognize as available to
us by which we may try (with some reasonable chance of success) to extend our knowledge and power.

One evident way for God to create creatures with this sort of choice is for him to create embodied creatures. An agent
has a body if there is a chunk of matter through which alone he or she can make a difference to the world, and through
which alone he or she can learn about the world. We humans are embodied because only through such a chunk (our
body) can we make a difference to other things: I can only move the furniture with my hands, or give you some
information by using my mouth to tell it to you or my hands to write it to you. I can only learn about the world by light
rays landing on my eyes or sound waves landing on my ears. And (barring a possibility to be discussed in Chapter 1 1) I
can only acquire knowledge or exercise power if my body is situated in a world governed by simple natural laws which
I can come to comprehend.

Good though the search for knowledge is, it can only be good in so far as knowledge is a good thing to have, and our search has some prospect of success. There is clearly
no point in searching for something you will not reach and is not worth having. Hence the famous remark of Lessing is a gross exaggeration: ‘If God were to hold out
enclosed in his right hand all truth, and in his left hand just the active search for truth, though with the condition that I should always err therein, and he should say to me:
Choose! I should humbly take his left hand and say: Father! Give me this one; absolute Truth belongs to Thee alone’ (G. E. Lessing, Erne Duplik , pt. 1).

ACTION

105

Only if, when I move my hands in certain ways, the furniture above them moves and stays put when I let go of it, can I
successfully move the furniture. And only if light travels in straight lines day after day will the visual images in my mind
and the beliefs about the world correlated with them caused by the light impinging on my eyes be reliable guides to
how things are in the world.

Now an embodied agent is one who starts with limited power and knowledge: he can move his limbs, and know about
his mental states. But if his body is situated in a world governed by simple natural laws, then he will note that moving
his limbs in certain ways leads to different mental states: putting my hand in a fire leads to pain and putting chocolate
in my mouth leads to pleasure. We come naturally to explain these correlations in terms of our bodies being situated in
a world of material objects which we can influence and which influence us. This interpretation of our situation comes
naturally. Further experience will add a litde to our knowledge unsought (if we are rational in our response to it). But
for significant expansion of knowledge, we need to choose to seek it actively. We can investigate the effects of making
various machines and pulling various levers, and test hypotheses about the causes of things. We can do this because
different states of the outside world have different effects on our bodies and so on our minds; and our different
actions through our bodies have different effects on the world, which we can study through the effects of the latter on
us. Embodiment in a world governed by simple laws of nature secures the goal of an initial limited repertoire of power
and knowledge, and a means (by discovering those laws) for extending it. God has this reason for making a universe
with simple laws of nature: to provide for embodied beings to have a limited repertoire of power and knowledge and a
means of extending it. But a universe governed by simple laws of nature will also have the beauty which our Universe
has, and which I set out in Chapter 3. In creating such a universe, God will fulfil two major purposes.

So God has a strong reason for creating creatures of limited power and knowledge: that they can choose whether or
not to grow therein. And he also has strong reason for creating creatures already having much power and
knowledge — in the goodness of power and knowledge. There is no particular best amount of power and knowledge in
a creature; and so God has reason for creating creatures of varying amounts of power and knowledge, and varying

106

THE GOOD GOALS OF CREATION

capacities for acquiring them — freely or spontaneously. All of which provides another reason for God to create
animals (with very limited power and knowledge, but some capacities to grow therein, both by learning passively, and
by actively seeking knowledge, yet spontaneously rather than freely) as well as the reason mentioned at the end of
Chapter 4: the goodness of having available for love those who cannot respond in quite the same way. There is too the
great goodness of the special kind of love that animals can have for humans, and the special kinds of cooperation
which can exist between animals and humans. And there is the great good of mere variety of kinds of power,
knowledge, and embodiment. Similar argument suggests that God has reason for creating non-embodied creatures
who have so much knowledge and power that they do not have much capacity to grow therein. Much Christian
tradition affirms that God has created creatures just like this — the angels. A principle powerfully influential in the
history of thought from Plato until at least the eighteenth century was the principle of plenitude: 88 nature or God would
create all the possible kinds of being.

In considering love in the last chapter, I considered only its involuntary aspects: desires to share things, including
beliefs; and desires that the beloved desire to share things including beliefs; and so on. But love also involves of course
the crucial component of action. It involves doing and seeking to do things with and for the beloved. And love is much
the better for being not merely mutual, but seeking (through a shared desire to do so) the well-being of others. Hence
the very great good of procreative love — bringing into being new agents who will be loved and live good lives. And
despite the value of animal love, obviously there is especial value in love which is freely given, in which the actions are
freely chosen. Is love the better if the loving actions are chosen, despite contrary temptation? Clearly it is good that the
lover sometimes persevere in dedicating himself to the beloved, despite contrary temptation. But

See A. O. Lovejoy, The Great Chain of Being (Harvard University Press, 1936) for the history of this principle. Aquinas, strongly influenced by this principle, claims that
diversity requires that there exist both immortal and mortal things; and also things that cannot cease to be good and things that can cease to be good; he holds that what can
happen, at some time will happen, and so there is bound to be an absence of good which, for Aquinas, constitutes the bad (see Summa Theologae , la. 48. 2). But of course
that is carrying the principle too far: the goodness of diversity as such provides inadequate reason for God to create what is bad.

ACTION

107

generous service in which there was little desire at all for the beloved could not be called love and, though a good
thing, would be significandy deficient. All desires for what is good are good, and the lack of desire for the well-being of
someone towards whom it is good that we perform loving actions is clearly a deficiency.

The Greatness of Being of Use

It is, this chapter has tried to bring out, a good thing to have power — to be able to make a difference to things for
good or ill. This I have called responsibility for things. But it is also a good thing to be of use, to help, to serve, either
through freely exercising power in the right way, or through doing it naturally and spontaneously, or even by being used
as the vehicle of a good purpose.

That helping is an immense good for the helper has always been difficult for humans to see, but it is especially hard for
twentieth-century secularized Western man to see. It is, however, something quite often near the surface of New
Testament writings, as I shall illustrate much more fully in Chapter 13. For the moment let me just quote St Paul's
farewell sermon to the church at Ephesus when he urged them ‘to remember the words of the Lord Jesus, how he
himself said, It is more blessed to give than to receive’. 89

We do not, most of us, think that most of the time. We think that our well-being consists only in the things that we
possess or the experiences we enjoy. Sometimes, true, all people find themselves in circumstances in which they ought
to give; alas, the starving appear on our doorstep and we ought to give them some of our wealth, perhaps something
large which will deprive us of future enjoyments. But that, the common thinking goes, is our misfortune, good for the
starving but bad for us. Life would have been better for us if they had not turned up on the doorstep. But what the
words of Christ say, taken literally, is ‘not so’. We are lucky that they turned up on the doorstep. It would have been our
misfortune if there had been no starving to whom to give; life would have been worse for us.

And even twentieth-century people can begin to see that sometimes: when they seek to help prisoners, not by giving
them more

20: 35.

108

THE GOOD GOALS OF CREATION

comfortable quarters, but by letting them help the handicapped; or when they pity rather than envy the ‘poor litde rich
girl’ who has everything and does nothing for anyone else. And one phenomenon prevalent in end-of-century Britain
draws this especially to our attention — the evil of unemployment. Because of our system of Social Security the
unemployed on the whole have enough money to live without too much discomfort; certainly they are a lot better off
than are many employed in Africa, Asia, or Victorian Britain. What is evil about unemployment is not so much any
resulting poverty but the uselessness of the unemployed. They often report feeling unvalued by society, of no use, ‘on
the scrap heap’. They rightly think it would be a good for them to contribute; but they cannot.

It is not only intentional actions freely chosen, but also ones performed involuntarily, which have good consequences
for others, which constitute a good for those who do them. If the unemployed were compelled to work for some
useful purpose, they would still — most of them — regard that as a good for them in comparison with being useless. Or,
if they would not so regard it I suggest that most of us who are employed, and not directly involved in their plight, can
see it as a good for them. Or consider the conscript killed in a just and ultimately successful war in defence of his
country against a tyrannous aggressor. Almost all peoples, apart from those of the Western world in our generation,
have recognized that dying for one's country is a great good for him who dies, even if he was conscripted . 90

And it is not only intentional actions but experiences undergone involuntarily (or involuntary curtailment of good
experiences, as

90 This good, others have recognized, exists as a this-worldly good, quite apart from any reward for patriotic behaviour which might accrue in the afterlife. The hope of such
reward was not a major motive among Romans and Greeks who died for their country. The doctrine of future life was far too vague among the pagans to exercise any
powerful general influence’ (W E. H. Lecky, History of European Morals from Augustus to Charlemagne (Longmans, Green, 1899), ii. 3 ). The Spartan and the Roman died for his
country because he loved it. The martyr's ecstasy of hope had no place in his dying hour. He gave up all he had, he closed his eyes, as he believed for ever, and he asked for
no reward in this world or in the next’ (ibid. i. 178). The well-known lines of Horace dulce et decorum est pro patria mori (‘it is sweet and proper to die for one's country’; Odes
3. 2. 13) were written by a man whose belief in personal immortality was negligible (see the famous ode 3. 30, in which he sees his ‘immortality 5 as consisting in his
subsequent reputation); and so those lines would seem to convey the view that dying for one's country was as such good for him who died. It was of course a Socratic view
that doing just acts was a good for him who does them (see Plato, Gorgias 479).

ACTION

109

by death) which have good consequences — so long as those experiences are closely connected with their
consequences — which constitute a good for him who has them (even if a lesser good than that of a free intentional
action causing those consequences, and a good often outweighed by the bad of the experience in question). Consider
someone hurt or killed in an accident, where the accident leads to some reform which prevents the occurrence of
similar accidents in future (e.g. someone killed in a rail crash which leads to the installation of a new system of railway
signalling which prevents similar accidents in future). His relatives often comment in such a situation that at any rate
the victim did not suffer or die in vain. They would have regarded it as a greater misfortune for the victim if his
suffering or death served no useful purpose. It is a good for us if our experiences are not wasted but are used for the
good of others, if they are the means of a benefit which would not have come to others without them, which will at
least in part compensate for those experiences. It follows from this insight that it is a blessing for a person if the
possibility of his suffering makes possible the good for others of having the free choice of hurting or harming him; and
if his actual suffering makes possible the good for others of feeling compassion for him, and of choosing to show or
not show sympathy or provide knowledge for others. Thus it is a good for the fawn caught in the thicket in the forest
fire that his suffering provides knowledge for the deer and other animals who see it to avoid the fire and deter their
other offspring from being caught in it. 91 I emphasize that I make this comment and similar comments in this and
other paragraphs without for the present raising the question who, if anyone, has the right to impose a bad state on an
unwilling agent Tor the sake of’ some good. We will come to that issue in Chapter 12. But I should make it clear here
that I am not saying that humans have any right to cause train crashes or forest fires of the kind just described. Nor am
I yet passing any judgements about whether the good is as great a good as the bad is bad. Nevertheless, I am claiming,
the supreme good of being of use is worth paying a lot to get. It is much better if the being-of-use is

Those familiar with recent philosophical writing on the problem of evil will realize that I choose the example of a fawn caught in a forest fire because of its prevalence in that
literature. It was put forward by William Rowe (“The Problem of Evil and Some Varieties of Atheism’, American Philosophical Quarterly, 16 (1979), 335—41 ) as an example of
apparently pointless evil. I shall be pointing out from time to time the good purposes which the fawn's suffering subserves.

110

THE GOOD GOALS OF CREATION

chosen voluntarily, but it is good even if it is not. Blessed is the man or woman whose life is of use.

Someone may object that the good is not (for example) dying for one's country, but knowingly dying for one's country
when one believes it good to do so — having the experience of ‘feeling good’ that one is sacrificing oneself. But that, for
reasons similar to those given in connection with similar claims earlier, cannot be right. It could only be good to have
the experience in question if one's beliefs were correct — we saw in Chapter 4 how a false belief can ruin the worth of a
mental state. There would be nothing good about believing one was dying for a good cause, when in fact it was a lousy
cause. And so the experience will only be good if what one believes is true, and one's beliefs include the belief that it is
good to die for one's country (which could occur even if one did not know it). What one rejoices in is that one's life is
of use, not that one knows that one's life is of use. 92 The good for the subject of being able to be of use to others (as
well as use his life in other good ways) does, however, make available to him the possibility of self-respect; he is in a
position (justifiably, not through self-deceit) to think well of himself as a person. Unjustified pride is of course a bad
thing, but proper pride is a good thing, and people who have it have a good worth more than many thrills of pleasure.

Actions which consist in bringing about what is bad or believed to be bad are as such themselves bad whether done
spontaneously or freely through yielding to temptation. Any failure to bring about what is good, when there is an
obligation to do good, is wrong. And actions may be wrong, because they involve a breach of obligation, even though
they do not consist in bringing about a bad state of affairs. When I talk with you, I implicitly accept an obligation not to
deceive you. So a lie is wrong, even if its consequences are good and even if the agent believed that its consequences
would be good. And being the victim of wrong is bad for the victim, not just because of the harm done but in so far as
it was done knowingly, or, worse, deliberately by an agent who sought the victim's harm. It is bad for someone to have
someone else seek their harm. No one likes

In the Old Testament, it is the Wisdom literature which saw most clearly this great good. The curse (in Wisd. 3: 11) for the ungodly is that their ‘works’ are ‘useless’; that, for
example (3: 16), ‘the children of adulterers shall not come to maturity’. The ungodly are compared to a ‘ship passing through the billowy water, whereof, when it is gone by,
there is no trace to be found’ (Wisd. 5: 10).

ACTION

111

to be hated — there would be something very wrong with us if we did — and seeking the harm of another is the natural
manifestation of hate.

A good God will seek to create agents not merely with good desires and beliefs but ones who have the opportunities to
do good actions and take those opportunities. If the good actions are not free, or do not involve a choice between the
perceived good and the temptation to do bad, God can ensure that not merely do the agents have the opportunity to
do good, but actually do it. And so he may well create many who do inevitably good actions.

God's Provision of Human Freedom

Clearly (if there is a God) he has created animals who (on the reasonable assumption that they do not have free will) do
inevitably many greatly good actions, such as feeding and caring for their young. Sometimes of course they do bad
actions, such as neglecting their young, and we shall need in due course to consider why God allows that. Clearly too (if
there is a God) he has created humans who make many serious responsible choices with all the characteristics
described in this chapter, and are (like animals) of use in many ways. But what is open to question is whether human
choices are free in the sense of not being fully predetermined by prior causes. This issue has of course been the subject
of (so far) unending philosophico-scientific controversy, to enter into which there simply is no space here. But suffice it
to make here two brief points.

The first is that it does often seem that it is up to us how we are to choose, and it is, as we have seen in Chapter 1, a
mark of rationality to believe that things are as they seem to be, in the absence of counter-evidence. The second is that
the mental life (of sensation, thought, desire, and purpose) is, evidently, so very different from normal physical events
in the inanimate world that the brain (in sustaining a mental life) must be very different from other physical systems
(which do not sustain mental life); and so there is not too much reason for supposing that any virtual determinism
which operates outside the brain operates within it. Hence there is not too much reason from science for supposing
that things are not as they

112

THE GOOD GOALS OF CREATION

seem to be in respect of our freedom of choice. But that does not of course setde the question finally; and the reader
will need to take into account the fuller philosophico-scientific arguments available in the literature. 93

That humans have free will (free will, that is, in the libertarian sense) is a doctrine which, I argued in Chapter 2, has
been strongly affirmed by a significant majority of the Christian tradition. Those who have affirmed it have seen it as a
central feature of the fact that humans are made fin the image of God’. Free will, they are telling us, is one of the best,
perhaps the best of all the characteristics with which God endowed humans. But the greatest glory of humans consists
in using it in the right way. Thereby humans will be conformed to God's ‘likeness’. In the fourth century, for example,
St Gregory of Nyssa wrote that ‘pre-eminent’ among all the facets of this image ‘is the fact that we are free from
necessity, and not in bondage to any natural power, but have decision in our own power as we please’. 94 In the eighth
century St John of Damascus wrote that ‘every man is said to be made in the image of God as regards the dignity of
his intellect and soul — as regards, that is to say, the quality in man that cannot be scrutinized or observed, is immortal
and endowed with free will, and in virtue of which he rules, begets and constructs’. 95 And at the end of the Eastern
patristic period, in the fourteenth century, St Gregory Palamas wrote that fit is not by virtue of natural qualities, but by
virtue of what one achieves through free choice that one is close to or distant from God’. 96 And St Gregory stressed, as
others did also, echoing the passage from Genesis with which this chapter begins, that part of the greatness of the
human soul is that it has ‘a capacity for sovereignty’: 97 the human soul ‘overlooks the universe and has all things in its
care’. 98 That is, in my terminology, the glory of humans is not just their very serious free will, but the responsibility for
so much which that free will involves.

For my own philosophico-scientific arguments in favour of human free will, see my The Involution of the Soul \ ch. 13 .

On the Making of Man , 16. 11.

St John of Damascus, ‘On the Virtues and Vices’, in G. E. H. Palmer, P. Sherrard, and K. Ware (ed. and trans.), Philokalia, ii (Faber & Faber, 1981), 341 .

St Gregory Palamas, ‘Topics of Natural and Theological Science’, in G. E. H. Palmer, P. Sherrard, and K. Ware (ed. and teans.), Philoka/ia , iv (Faber & Faber, 1995), 382 .

ACTION

113

Because of the weight of Christian tradition in its favour, and the absence of good philosophico-scientific arguments to
the contrary, I am therefore taking the doctrine that humans have free will for granted. I have argued that it is a good
thing that we should have it with the resulting responsibility for ourselves and others. It thus forms a central part of the
package of Christian doctrine in terms of which I seek to explain why God should allow the world's bad states to
occur. Given the omnipotence of God, humans have free will only in so far as God continually sustains it in them and
allows their actions to have their intended consequences. But just as it is good that human parents should allow their
children to make choices which are not good, and it is the mark of a good parent that she should allow her children, as
they become older, to rebel; so surely a good God will do the same. An objector may, however, question whether, if
there is a God, he allows humans too much free will and responsibility; I shall come to such quantitive issues in
Chapter 13.

The great good of so much free will and often cooperative human action has gone with the bad of so much negligence
of ways to prevent harm, and so much deliberate infliction of harm (often cooperatively). Our bad actions, like our
good actions (especially if done in cooperation), have effects in distant places and distant centuries. The Holocaust was
made possible by the continued denigration of Jews over many centuries, by professed Christians of Central Europe. I
need to argue in due course for the greater good which allowing (and when we are dealing with free human actions, it
is only allowing, not causing) such horrible things to occur makes possible. But it is right to start by dwelling at some
length on the very great and (as far as much of this chapter is concerned) manifest goods of the created order.

The Influence of Angels and Ancestors

There are two further Christian views about the range of responsibility of free creatures for the state of the world,
suggesting that it is much greater than may at first appear in the light of mere secular considerations, to which I drew
attention in Chapter 2.

The first view is that there are angels, non-embodied creatures who have some influence for good or ill over the world.
Given that

114

THE GOOD GOALS OF CREATION

there is a God, this ancient view has some a priori probability (reinforced by any probability it might gain, if seen as
part of the Christian revelation). There are many, many kinds of good state. Embodied humans are not the only kinds
of rational agent with free will there could be. God might well create non-embodied free agents; and some of them
might indeed choose the bad. We may perhaps regard the initial choice of character by the angels as one extended in
time, and, given that, if it is good that God should give us the ultimate choice over the period of our lives on Earth of
being allowed to fix our characters beyond further change, it would seem to be similarly good that God should give to
angels also the ultimate choice of being allowed to fix their characters. And he might well also, in giving them that
initial choice, have promised them temporary and limited power over the world when they had fixed their characters.
That would have given a deeper significance to their choice than it would otherwise have, for they would know that in
choosing the bad, they would be choosing to exert a bad influence on the world. If freedom and responsibility are good
things, it is good that there be angels who have it, as well as humans. The power of the angels might have been
exercised by their influence on the structure of the world which God created (e.g. ensuring that natural processes
sometimes produced incurable pain) rather than by intervening in an already created order. However, the Christian
tradition has on the whole seen the influence of bad angels (and so of the chief bad angel, the Devil) as operative
mainly in tempting humans to wrongdoing, rather than in causing suffering. And in view of the limited place of such
demons in the Christian tradition, it would seem ad hoc to attempt to explain too much of the bad in the world as due
to their operation; there is not enough independent evidence to suppose that any bad angels have very much power.
Analogously there is not enough independent evidence to suppose that any good angels have very much power. In any
case, since angels could only choose the bad if they were tempted so to do, being already subject to bad desires, the bad
must have pre-existed any bad choice by angels.

Then, secondly, there is the view that human free choice has more distant effects than we realize. There is in particular
the doctrine that it was through the knowingly wrong choice of the first human being, i.e. the earliest of our ancestors
with free will and moral awareness, traditionally called Adam, that natural evil and

ACTION

115

death and at any rate much of the proneness to moral evil came to the human race. We saw in Chapter 2 that this is
most plausibly expressed as the view that, but for the sin of Adam, God would have conferred on humans an
immunity to suffering, to bad desires and other natural evil, and to death.

Now clearly (whether he lived many thousands or millions of years ago) there must have been a first human being in
the sense of an earliest of our ancestors who had moral awareness and (given that we have free will) also free will. For
the primitive organic material out of which humans and animals developed did not have those characteristics, and, I
am assuming, we do. And plausibly enough — at any rate given the view of Adam as a weak character which Irenaeus
and other early theologians advocated and which is far more plausible given our knowledge of evolutionary history
than the rival view — the first human yielded at some time to his bad desires and so was also the first sinner. (I write
this without any implication that the sin consisted in eating an apple in any literal sense.)

It may be that Adam's bad action had more distant effects and bigger effects than at first appears, because God
allowed human mortality and the extent of human suffering and proneness to sin to be affected by that choice. But as
with the angels, so with Adam: Adam could not have sinned without already having some bad desires to which he
yielded — and bad desires are themselves a natural evil. Adam must have inherited from his animal ancestors such
desires as the desire to eat more than his share of the available food and the desire to order others about. What was
different about Adam is that because he had moral awareness and free will (that was what his having a human soul
amounted to), such actions became sinful, and because of his moral awareness he came to see such actions as wrong
and (given that he had also some awareness of God) as sinful. By inheriting, genetically, his desires and his free will;
and by inheriting, no doubt in part culturally, his moral awareness, we later humans inherited that sinfulness. But we
would have inherited the sinfulness anyway, whether or not Adam had sinned, but for divine intervention.

As we have seen, the traditional account of the Fall claims that God would have saved us from death and suffering
(and on some versions from sinfulness also), but for the sin of Adam. Perhaps so. As I am arguing throughout this
book, although the bad in the

116

THE GOOD GOALS OF CREATION

world serves good purposes, God would have been no less than perfecdy good if he had produced a world with a lot
less bad and so a lot less of that good for which the bad is needed. And maybe God allowed which of the worlds he
would produce (a world without human death and much human suffering; or a world with these and also the other
good states which they make possible) to depend on the choice for good or ill of the first human being. But there must
be much other good, than the good of Adam having the responsibility for the future of so many millions of
subsequent humans, served by our subjection to suffering. For good though human responsibility is, it would surely
have been bad if God gave to one human being the choice of whether all subsequent humans suffered or not, unless
much other good was served by that suffering." This would be especially bad since, given the implausibility of the
‘original righteousness’ account of Adam , 100 he would not have understood much of what would have followed from
that choice. If Adam's choice made all the difference, God would only have been good in allowing him to have that
choice if overall the alternative consequences of his actions did not differ enormously in their goodness. There must be
a lot of other good served by our suffering than the good of Adam's responsibility for future generations. For this
reason the Fall must not be given too central a place in theodicy. My own theodicy will be concerned to bring out how
suffering and sinfulness make possible good states for those who suffer and sin and their immediate victims, rather
than give too much importance to the role of Adam or the Devil; I believe that I am following the best traditions of
earlier Christian thought in this respect.

99 And plausibly it would have been almost equally bad if God had given to a human the choice of whether all subsequent humans were mortal or immortal — though the issues
here are different, if death is not itself a bad state but merely the end of a good state, as I shall be arguing.

100 See pp. 39 and 101.

6 Worship

And there came one of the seven angels . . . And he carried me away in the Spirit to a mountain great and high,
and shewed me the holy city Jerusalem, coming down out of heaven from God, having the glory of God . . .
And I saw no temple therein: for the Lord God the Almighty, and the Lamb, are the temple thereof. And the city
hath no need of the sun, neither of the moon, to shine upon it: for the glory of God did lighten it, and the lamp
thereof is the Lamb. And the nations shall walk amidst the light thereof: and the kings of the earth do bring their
glory into it . . . And he shewed me a river of water of life, bright as crystal, proceeding out of the throne of God
and of the Lamb, in the midst of the street thereof. And on this side of the river and on that was the tree of life,
bearing twelve manner of fruits, yielding its fruit every month: and the leaves of the tree were for the healing of
the nations. And there shall be no curse any more: and the throne of God and of the Lamb shall be therein: and
his servants shall do him service; and they shall see his face; and his name shall be on their foreheads. And there
shall be night no more; and they need no light of lamp, neither light of sun; for the Lord God shall give them
light and they shall reign for ever and ever.

(Revelation 21: 9-22: 5)

The good states of which I have written so far are good whether or not there is a God. But there are some states which
are good if and only if there is a God. And most of them fall under the central heading of a right relation to God
himself. A good God, for his sake and for theirs, will seek to have creatures rightly related to himself. He will seek for
those creatures who have the capacity to do so to know him, interact with him, and love him.

We saw in Chapter 4 what a good thing knowledge (in the objective internalist sense) is, propositional knowledge,
knowledge of

118

THE GOOD GOALS OF CREATION

facts, that is. All knowledge is good, but especially knowledge of deep truths about the Universe and our place in it,
who or what is the source of our being, and moral truths about our duties and the good actions beyond duty which we
can do. Humans and maybe rational beings elsewhere in the Universe (if any such there are) are capable of
understanding true answers on these issues — at any rate to some degree. So if God is the creator and sustainer of the
Universe, he will want us to have such knowledge, and so to know (for our sake, of course; not for his sake) that
among our moral duties are the duty to express our gratitude to God and seek to do his will. For, as we have noted,
there is a duty to please benefactors. God is our supreme benefactor: we owe our existence from moment to moment,
and our powers and pleasures, our knowledge and desires, to his sustaining power. And everything that everyone else
does for us they can do only because God sustains in them their power. We owe him so much by way of expression of
gratitude and service; and, as was noted in the last chapter, it is supererogatorily good to do more for benefactors than
duty requires. And since God, seeking our well-being and theirs, will want us to be good, to fulfil our duties to others,
and to do things for them and ourselves beyond the call of duty, it will become doubly our duty and doubly
good to do so. 101

All human wrongdoing to other creatures is also wronging God. For, first, anyone who misuses a gift from a
benefactor wrongs the benefactor. God gives us life and all our powers and knowledge. If we use them to hurt
someone else, we wrong God also. And secondly, just as anyone who wrongs a child wrongs the parent who brought
him into existence, nourishes, educates, and loves him, so wronging another creature is wronging God whose creature
they are. Wronging God is sin; and so all wrongdoing is sin.

But God is not just our benefactor deserving our grateful service. I reflected in Chapter 4 on the goodness of having
and expressing respect for the powerful, wise, and good. It follows from God's omnipotence and perfect goodness
that he is holy, so other and different from anything else that a truly awesome reverence is the only proper reaction to
him. And God will want us for our sake to respond in the proper way to his own ultimacy.

101 For fuller justification of the claims of this paragraph, see my Responsibility and Atonement (Clarendon Press, 1989), ch. 8 .

WORSHIP

119

God will, for these and other reasons to which I will come shordy, want us to know that he exists. He could do that by
showing us that some argument for his existence from evident features of the world is deductively valid or gives high
inductive probability to its conclusion. Or he could give us a very powerful experience of his presence. Although the
human concepts in terms of which we attempt to describe and recognize God are inevitably hopelessly inadequate for
the purpose, on the traditional Western view with which we are operating, they give us some idea of God. God is
powerful, rather than weak; wise, rather than foolish; good, rather than bad. But also on that view they are just
adequate enough for us who use them to realize how inadequate they are. To quote those haunting words of Kant:
‘Unconditioned necessity, which we so indispensably require as the last bearer of all things, is for human reason the
veritable abyss. Eternity itself, in all its terrible sublimity, as depicted by a Haller, is far from making the same
overwhelming impression on the mind; for it only measures the duration of things, it does not support them .’ 102 It is
natural to suppose that the inadequacy of our ordinary categories for talk about God arises from the fact that they have
their primary application for talking about ordinary mundane beings. We learn the contrast between ‘wise’ and ‘foolish’
by having examples of wise and foolish humans pointed out to us; and then we are told that God is more like the
former than the latter (in the respect in which the former are like each other and unlike the latter). But if God gave us a
more direct awareness of himself, we could devise words and thus designate categories for describing the God of
whom we are aware in experience, which would capture more nearly what God is like. God will surely want to give us
this knowledge.

As our creator, God will seek to interact with us. He will want us to feel his presence, to tell him things and ask him to
do things; and he will want to tell us things (to be the source, and known to be the source, of all the deep truths to
which I have referred), and to do things with us, to cooperate with us in producing further good. No parent would be a
good parent who did not seek all this for their children; and God, as so much more the source of our being than our
earthly parents, must seek all this for us. So he will manifest his presence to us, not merely give us the knowledge that
he is present.

120

THE GOOD GOALS OF CREATION

He will persuade us to tell him things (even if he knows them already; many a parent knows already what his child tells
him, but wants the child to tell him all the same). He will want us to ask him for good things for ourselves and for
others, which he will bring about if and only if we ask; for in that way, the good which he brings about is better for our
involvement in its production. And God would wish to involve us in his projects, and his concern with each of us
would be shown by giving us different jobs to do, by tailoring jobs to fit us rather than just saying ‘Do something
useful’.

When we do wrong, God will want us to hear his angry rebuke; when we apologize, God will want us to hear his
message of pardon. And as a loving parent, he will want us to love him; and he will cultivate in us the beliefs about him
and desire to interact with him which make this possible.

While all loving interaction, alias friendship, is good, such interaction with those with whom we are deeply involved for
other reasons is especially good. Friendship with our benefactors and beneficiaries and so with parents and children is
especially good because it consecrates with liking (i.e. saves from formality) relations of deep significance. And
friendship with good people is especially good. This is because it involves sharing good thoughts and good activities,
and liking those who are especially worthy of being liked because they do good actions and have good characters, and
so will involve having the right attitude towards actions and characters. (I do not of course wish to imply that we
should seek friendship only with the already good, but only that since friendship is better for being based on fully
correct values, we should aim at turning any friendship into such a friendship.) All friendships, and especially those of
the kinds just discussed, are intrinsically valuable. Friendships with the good, and especially with good parents
deserving respect and gratitude, are also obviously enormously valuable instrumentally. The good influence of good
parents is very great. The good influence of a good God would be enormous.

Now the Christian tradition teaches two doctrines relevant to such divine interaction with us. The first is that God
does now have limited interaction with many Christians and others. And the second is that God will have full
interaction with many of us after death. As regards his present limited interaction with us, God does, it is claimed, help
some of us to see the force of arguments to his existence, give some of us experience of his presence, make known

WORSHIP

121

to some of us our Vocations’. And certainly to some of us it seems that this is the case. If there is a God, it seems
reasonable to suppose that our apparent experiences of him are genuine, and that often what he seems to be telling us
he is telling us; and in so far as our arguments have force, it is God who is helping us to see that. The claim continues
that humans have a duty to interact with God in prayer, and to make petitionary prayer for the removal of many bad
states and the coming about of many good states; and that God answers many of the prayers of those who ask (in the
way most expedient). God does not answer all prayers, and his goodness would not lead us to suppose that he
would . 103 Good benefactors do not answer prayers for what is bad; and they only answer prayers for what is not bad
only in a certain respect, in that respect But God does answer some prayers, the claim is, in the sense that he brings
about some state of affairs prayed for which would not otherwise have occurred. Some people cured from some
disease would not have been cured but for prayer, and some people not cured would have been cured if others had
bothered to pray for their cure with sufficient devotion. If human responsibility is good, then this extension to it — of
exerting influence on (though not of course compelling) God to change things — would surely be also good.

It has been central to the Christian tradition since its beginning that God is an interactive God, guiding us and
responding to our good and bad deeds individually. It has been so much taken for granted by the tradition that it has
never been sufficiently a matter of controversy (before the last two or three centuries) to require formulation as a
doctrine. If the investigator finds that science or philosophy has counter-evidence to this claim, she will have to take it
into account; but in so far as the evidence is from revelation, it is very strongly in its favour.

There were unanswered prayers both of St Paul (2 Cor. 12: 7-9) and of Jesus himself (Mark 14: 36). On the other hand the Gospels have sayings seeming to say that all
prayers, at any rate of Christians, will be answered (e.g. Matt. 21: 22). The usual Christian compromise is to suppose that the Gospel overall is saying only that prayers for
things which are not contrary to the will of God will be answered. They may be either for states neither good nor bad, or for good states (or the removal of bad states) but
ones which God will only bring about if asked — because he wants us to be involved in bringing them about.

122

THE GOOD GOALS OF CREATION

The Mechanism of Providence

The good states — knowledge, good desires, efficacious desires, etc. — with which previous chapters were concerned
are ones given to all or to large groups of creatures. Given that there is a God, they thus manifest what is known as his
‘general providence’. (Apart from the possible influence of angels and of Adam) they are goods arising from the
general structure of the world, the natural order of things; which, if there is a God, clearly he brings about.

God's dealings with particular individuals, however, in response to their particular needs and requests, not in
accordance with any general formula, manifest God's ‘special providence’. This involves his intervening in the natural
order of things. The ‘natural order’ of things is simply their regular behaviour in accordance with the laws of nature. If
this regular behaviour, the ‘laws of nature’, are deterministic, any intervention by God will involve his temporarily
suspending the laws of nature, violating them. But if this regular behaviour is not fully determined (if the laws of nature
merely make some future events physically probable and others physically improbable, but do not determine what will
happen), then God's intervention will simply be a matter of inclining things to behave on a certain occasion in this way
rather than that, which will be compatible with the probabilistic patterns of behaviour which humans may discover and
formulate as ‘laws’. Yet even if compatible with the ‘laws’, some divine interventions could cause states of affairs very
improbable given those ‘laws’.

Quantum theory indicates that the most fundamental laws of nature, the laws governing the behaviour of very small-
scale particles, are probabilistic, i.e. indeterministic; but that, in general, small-scale indeterminacies cancel out on the
large scale, leading to virtually deterministic behaviour of the medium-sized objects with which we interact — tables and
chairs, trees and persons. The human brain may well be an exception — here small-scale indeterminism may well lead to
large-scale indeterminism. Hence the possibility that our free actions , 104 and any effects of divine intervention in our
mental life, are not merely compatible with the laws of nature, but not improbable given those laws. But other divine

theory makes room for free will,

ly The "Evolution of the Soul, re

edn. (Clarendon Press, 1997), ch.

WORSHIP

123

interventions in the world, God moving material bodies over several metres, for example, would require the
occurrence of events immensely improbable given natural laws, which we may call ‘quasi-violations’ of natural laws. If
we are to know that some event is or is not a quasi-violation, we need to know not merely what are the laws of nature
but what was the previous state of the world. Suppose that the desk seems suddenly to move several metres from here
to there before our very eyes. We need to know that there are no ropes pulling it, that no magnets are at work, that it
really is a desk and not a hologram, that we are not subject to a sudden hallucination, etc. Occasionally we are in a
position to know this. But with respect to events crucial for human well-being (health, crops flourishing, accidents,
defence against enemies, success in business), we would be very seldom indeed in any position to know that some
event which constituted in a wide sense what was prayed for did or did not occur in accordance with natural laws. This
is because all events, and particularly events of this kind, are such that which subsequent events will follow earlier
events in accordance with laws of nature depends on a whole multitude of interacting earlier events (many of them
very small-scale ones).

We are normally only aware of some of these earlier events, and when some subsequent event occurs we suppose it
caused by the previous events of which we were aware, combined with whatever other events would be sufficient in
accordance with laws of nature to produce the subsequent event. That is, we ‘read back’ into nature causes sufficient to
produce whatever we observe. We are almost always able to do this, and very seldom would we be in a position to say
that some event could not be explained by ordinary causes operating in accordance with known natural laws. Someone
recovers from cancer. We would hardly normally be in a position to know enough about the previous state of their
body to know that natural laws would not lead from the latter as cause to the recovery as effect. Pharaoh's army is
drowned by the Red Sea suddenly rushing in and covering a previously shallow crossing. Natural causes could well
have combined with the state of wind and rain at that place on the previous day to cause Pharaoh's army to drown.
Alternatively, God may be bending nature to answer prayers and thus interact with individuals and groups all the time;
we would not normally know. Just occasionally we are in a position to know that a violation or quasi-violation has or
has not

124

THE GOOD GOALS OF CREATION

occurred . 105 And maybe one day we will be in a position to know this with respect to a lot more events, and if so maybe
there would be evidence to show that God very seldom intervenes in nature, and if so that would show something very
much in error in the Christian tradition. But I cannot see that present evidence shows that future discoveries are more
likely to show that than to show the opposite. And so I suggest that the doctrine that God intervenes in the natural
order from time to time to answer prayers in the ways most expedient for us is not at odds with our knowledge of the
natural world, and so to be adopted on grounds of revelation.

Yet if God answered all prayers for the removal of bad states of affairs promptly and predictably, that would become
evident. Clearly that does not happen. Maybe God allows some bad states to continue in order to allow people the
choice of whether to persist in prayer and allow more people to be involved in the prayer. Yet some bad states seem to
be so evidently unaffected by much continuing prayer that natural evil cannot in general be accounted for by the
absence of petitionary prayer. We need other reasons why God should often not answer such prayers, in terms of some
good states (other than people having the free choice of whether or not to pray for the removal of bad states)
promoted by allowing the bad states to continue.

Life After Death

The second doctrine is that God will take many of us after death, and — if the world ends before some of us
die — many of us before dying, into his more evident presence, and we will be able to render proper worship and other
service to him then. The last book of the Bible, from which the quotation at the beginning of this chapter is taken,
describes the saints as worshipping God, and ‘reigning’, that is, exercising power with God for making the world or
maybe new worlds ever better. The Catholic tradition has represented the state

For arguments about what would be evidence for and against such violations, see the papers in R. Swinburne (ed.), Miracles (Macmillan, 1989) . For examples of a few
modern cases where there is quite a balance of evidence in favour of the occurrence of a violation or quasi-violation, see D. Hickey and G. Smith, Miracle (Hodder &
Stoughton, 1978) or Rex Gardiner, Healing Miracles (Darton, Longman & Todd, 1986) .

WORSHIP

125

of the inhabitant of Heaven as enjoying the ‘Beatific Vision’, a state of contemplation of God, an expression which
seems to do less than justice to the activity which is so good a feature of the life of humans and other rational beings.
The Orthodox tradition has described it as 00tofcjbic;; which, translated literally, means ‘deification’. For most of the
Fathers this meant simply ‘enjoying the divine relation of Son to Father, sharing the divine life’. 106 The blessed, although
not themselves divine, interact with the Father as closely as does the Son, and so are involved in the work of God as
well as the adoration of God. Since the wider Christian tradition has in general held — although not describing it as
‘deification’ — that this latter is also involved in the life of Heaven, I shall in future understand the ‘Beatific Vision’ as
including it.

Who will enjoy the life of Heaven is something about which the Christian tradition has been far from clear. Christian
believers who have lived on Earth a good life which ended with a good death are clearly included. But ‘believers’ has
been traditionally thought to include also those who have been baptized and died in infancy. It has been thought to
include also the Jews who lived good lives before Christ; and more generally the good ancestors of the Jewish people,
including Adam. Some early theologians, like almost all modern ones, have thought that all who have not heard the
Christian Gospel presented with sufficient plausibility, but who have lived good lives and made themselves so sensitive
to truth and goodness as naturally to worship God in Christ when after death they hear the Gospel, will also enjoy the
life of Heaven. A ‘good life’ has not usually been thought of as ‘many good works’, but a life which ended with a
readiness to do good works, even if litde time remained in which to do them. But while some Christian theologians of
the present century have been ‘universalists’, that is have held that necessarily all humans will eventually be ‘saved’, i.e.
go to Heaven, that position was virtually unknown in earlier times. That humans who fall outside the limits stated
above in their widest sense will not be saved has been the traditional Christian view (although it was always allowed
that just maybe in fact every human falls within the limits in their widest sense). 107

Rowan Williams, The Wound of Knowledge, 2nd edn. (Darton, Longman & Todd, 1990), 51 .

For a more detailed account of what the Christian tradition has supposed about the life of Heaven a:
afterlife, see Responsibility and Atonement \ ch. 12 .

and whc

aid enjoy it, and the different fates of different groups in the

126

THE GOOD GOALS OF CREATION

Both of these doctrines, that God does and will interact with us in these ways, to be expected in virtue of his goodness
and also further evidenced to the extent to which there is independent evidence for the Christian revelation, do raise
obvious problems: why is God's present interaction with us so limited, and why is his future interaction not provided
for all? But we need to reflect on whether and why the limited interactions are bad states of affairs rather than mere
absences of a good. For, as we have seen, there is no end to good states, and not even an all-good and all-powerful
being could produce all the good states that there are. The limitedness of God's interaction in the present would,
however, seem to be a bad state. It is, first, a bad state for those humans who desire to interact with God and desire to
interact with him more fully. For then they have a frustrated desire, and that is a bad thing, I have argued. If humans
long to have a deep experience of God, to interact with him in prayer, or at any rate to know that he is there, and do
not — here is something bad. To the extent to which humans want God and do not have him there is something bad.
But mere ignorance of God's existence is a bad thing for quite a different reason. Humans have, as I have noted, a duty
to worship and serve God; and if they are ignorant of his existence, they cannot fulfil those duties, and an inability to
fulfil a duty is clearly a bad state. In these ways and to these extents lack of knowledge of and interaction with God in
the present is a bad state for humans; and we shall need in due course to see what greater good that ignorance and
darkness serves.

Would it be a bad state if God's future interaction is not provided for all humans, if not all humans will enjoy the
Beatific Vision? Now that again will only be a bad state (as opposed to the absence of a good) if those deprived of that
vision desire it all the same. If some do not desire it, it would not be a bad thing if they do not have it. As we have seen,
the life of Heaven consists in adoration of the good and in pursuit of the good. It therefore involves wanting and
seeking all that God wants and seeks — wanting the wrongdoer to be penitent, and wanting the good to be generous
with their lives, and wanting to worship God for his goodness. It would indeed be a bad state if anyone who desired
these things (including, if they have done wrong, their own penitence) were deprived of the life of Heaven.

Why should anyone be in such a condition of not desiring the good? Either because God has allowed them to get into
this condition,

WORSHIP

127

even though once they desired the good, or because they never had a desire for the good in the first place. I pointed
out in Chapter 5 that humans are so made that their good choices on one occasion make good choices easier next
time. We are so made that we gradually form our characters — either so that we become naturally good (prone to do
the good, as soon as we are righdy informed wherein it consists), or so that we naturally yield to bad desires such that
they acquire permanent mastery over us. It is good that God should allow people the choice of forming their
characters in such a way as not to be open to future change. For if God refused to allow someone to develop an
irreformably bad character, that would be depriving her of an ultimate choice of the sort of person she is to be. If God
always left the bad open to good influences, that would be refusing to recognize an ultimate choice by an independent
moral agent. It would be like a jilted lover pestering the beloved on and on, not recognizing her right to say a final no.
But if someone does form their character in such a way as to be unalterably bad and if that involved their having no
residual desire for the good which they cannot choose, they do not then desire the Vision of God; and so it is not a bad
thing that they do not get it. (Whether the continued choice of the bad does involve elimination of any residual desire
for the good is an issue to which I shall come in Chapters 8 and 11.)

There may also be those who never had a desire for the good (i.e. the morally good) in the first place; and there would
be nothing bad in them not obtaining the Vision of God. Many medievals believed that the unbaptized who die as
babies are in this category. They have not committed actual sin, and so they are not punished, but they do not enjoy the
Vision of God. They go to an intermediate Limbo (more technically, the Umbus puerorum) where their enjoyments are
on a lower level. To all appearances animals also lack moral concepts, and so cannot desire the Beatific Vision; and so
their not getting it is not a bad thing. Indeed it rather looks as if animals in old age have no desire to continue to live at
all, and so if they have no life after death, that too requires no theodicy.

Christian tradition has, however, taught not merely that the incorrigibly bad are deprived permanendy of the Vision of
God, but that this is a punishment for their sins {poena damni , the penalty of the loss of God) and also that they suffer
further felt punishment {poena sensus), i.e. punishment involved in frustration of desires.

128

THE GOOD GOALS OF CREATION

That is a bad state, not merely the absence of a good one, and so requires a theodicy. I shall come to such a theodicy in
Chapter 11.

While noting the need for theodicy in respect of the temporary lack of the Beatific Vision for those who desire it, and
for the felt punishment of those who permanently do not, let us return, at the end of Part II, to reflect not merely on
the goodness of an Earth so evidently good in the ways described in earlier chapters, but on the glorious state which,
the Christian revelation affirms, God will give to those who desire it. Blessedness with the saints in the life of God
himself, in adoration of the good and ever promoting that good further, is, according to the Christian revelation, what
God, our loving parent, had in mind in creating us.

Ill The Necessary Evils

This page intentionally left blank

7 The Fact of Moral Evil; and Free Will

[The Lord] himself made man from the beginning and left him in the hand of his own counsel. If thou wilt, thou shalt
keep the commandments; and to perform faithfulness is of thine own good pleasure. He hath set fire and water before
thee: thou shalt stretch forth thine hand unto whichsoever thou wilt. Before man is life and death; and whichsoever he
liketh, it shall be given him.

(Ecclesiasticus 15: 14—17)

The Logical Straitjacket

God will seek to provide all the good things and none of the bad things described in Part II. But he cannot — for
reasons of logic. For, as simple non-religious examples will make evident, some good states are logically incompatible
with each other. It is good for John to be married monogamously to Mary for all his married life and good for John to
be married monogamously to her sister Ann for all his married life, but it is not logically possible that he be in both
these good states. It is good for Bill to be the President of the United States in AD 2000, to have the great responsibility
of government; and good for Bob to be President of the United States in AD 2000. But they cannot both be President
of the United States at the same time. And some good states are not compatible with the absence of certain bad states.
It is good to have games of football, in which teams develop their ball skill and team skill in such a way as to defeat
other teams. And it is good that all the teams want to win, and so are enthusiastic, and happy when they win. But if a
winning team is happy because its desire to win is being fulfilled, a losing team will be unhappy because its desire to
win is being

132

THE NECESSARY EVILS

frustrated. Arsenal's team and Liverpool's team cannot both win, and Arsenal cannot be happy at winning unless
Liverpool are unhappy at losing. And not even God who can do anything logically possible can bring about both of
two incompatible good states; or a good state without the bad state, when the good state entails the bad state.

Actions may be logically impossible for reasons other than that they consist in bringing about logically impossible
states. There is no logical impossibility in there being an uncaused state, but it is logically impossible for anyone to
bring about an uncaused state; and there is no logical impossibility in there being a state of affairs uncaused by me, but
it is not logically possible for me (though it is logically possible for you) to bring it about. God cannot do what is
logically impossible for him to do — whatever the reason for that logical impossibility.

People sometimes think that a really powerful God ought to be able to do the logically impossible — bring it about that
John is monogamously married both to Ann and to her sister Mary for all John's married life. But to say that some
‘action’ is logically impossible is just to say that what appears to be a description of an action does not make ultimate
sense, i.e. there is nothing which would constitute doing such an action. It is obvious that there is no action of making
my desk both exist and not exist at the same time: there is nothing which would amount to doing such an action.
Whatever I or anyone else did, whether in consequence my desk existed or not, there is nothing which would
constitute making it exist and not exist at the same time. There are other purported descriptions of actions which look
as if they describe actions which someone (at any rate someone sufficiently powerful) might do, but do not really. The
description in that case does not, I shall say, make ultimate sense. If we investigate more fully what it would be like for
such a description to hold — e.g. make John monogamously married both to Ann and to her sister Mary for all his
married life — we realize that there is nothing which would constitute doing such an action. The logically impossible
action is not an action which is very difficult to do, but a purported description of an action which does not describe an
action at all. But, having made that point, we can make the original point (without now, I hope, misleading) by saying,
God cannot do the logically impossible.

It follows from that, I shall show in the next four chapters, that God cannot bring about (or at any rate cannot
guarantee to bring

THE FACT OF MORAL EVIL

133

about) many of the good things described in Part II at the same time as many other good things, nor many of the good
things without many bad things as well. I shall consider in turn the various bad things of this world, and show how
their actual occurrence or the unprevented possibility of their occurrence is required for the realization of the various
good states listed in Part II. The most obvious example of this logical straitjacket to which even God is subject is that
he cannot give us very serious free will, i.e. the free will to choose between good and wrong, without the natural
possibility (unprevented by God) that we will do wrong.

The Free-Will Defence

So I begin with one of the obvious ills of this world — the core moral evil of agents making wrong choices. I argued in
Chapter 5 that it is good for agents to have very serious libertarian free will — libertarian, in that it involves freedom to
act in a way not fully determined by causes; and very serious, in that it involves freedom to choose between the good
and the bad, including the wrong. It would seem logically impossible for God to give agents this freedom (to make it
up to them how they will choose) without the probability of them making some wrong choices. The more agents who
have this freedom, the more such choices they have, and the greater the temptation to wrongdoing, the more wrong
choices there will probably be. (Just how probable will depend on the strengths of their wrong desires — on which see
later.) If humans have libertarian free will (as I am assuming; see Chapter 5), then the unprevented possibility of human
wrong choices makes possible the great good of human free choice between right and wrong.

Plantinga's Complication

Alvin Plantinga is a firm advocate of this position, the Tree-will defence’ to the objection to theism from the existence
of moral evil. 108 But he has raised a complication for his own position. He

For Plantinga's initial discussion
developed earlier in that book; a

}f the problem of evil, s

•e of Necessity (Clarendon Press, 1974), ch. 9 ,

’ and Evil (George Allen & Unwin, 1974), pt. 1

134

THE NECESSARY EVILS

holds that, although God could not make people choose freely as they do, he foreknows how they will choose freely;
not just ‘foreknows’ in the sense of ‘is aware of their present desires and beliefs, and so knows how likely it is that they
will do the good as they see it or alternatively capitulate to desire and do wrong’, but knows necessarily (namely,
without the possibility of mistake) exacdy what they will do. Plantinga holds this because he holds to a strong
traditional view of God's essential omniscience, as God knowing all true propositions; and combines this with the view
that God is not timeless but everlasting, and so knows now how people will choose tomorrow, that is has
foreknowledge of their free choices. Such omniscience on this view is essential to God: he cannot but know all this, or
he would not be God. Plantinga also holds that God's omniscience includes (essentially) ‘middle knowledge’, that is,
knowledge not merely of how actual beings will choose (freely); but also knowledge in advance, of how all possible
beings would choose (freely) in all possible circumstances if he were to create them. God in the beginning, at stage one,
knew all the worlds which he could create, in the sense of which free beings he could create and in which
circumstances governed by which (deterministic or probabilistic) natural laws he could place them; and also which free
choices each of them would make in the different circumstances. He then used this middle knowledge to decide to
create one such world — that is, to create some of the possible beings and put them in some of the possible
circumstances. Then in virtue of his decision, at stage two he acquired ordinary foreknowledge of what these beings
will choose freely; and his middle knowledge became knowledge of what would have happened if he had acted
otherwise . 109 God knows what I would have done if I had fought in the Second World War, been taken prisoner by the
Japanese, and been offered a bribe

109 The doctrine of ‘middle knowledge’ was first put forward (under that name — but see my Additional Note 5) by Luis de Molina in pt. IV of his Concordia (1588). For an
excellent introduction to and translation of the relevant parts of this, see L. de Molina, Of Divine Foreknowledge, trans. A. J. Freddoso (Cornell University Press, 1988). If one
thinks of God as outside time, then these ‘stages’ in God's knowledge acquisition are not temporal stages, but ‘logical’ stages in God's timeless knowledge possession; having
‘ordinary foreknowledge’ of how beings will choose is to be interpreted as having ordinary knowledge of how they will choose at times future relative to some time-bound
human observer.

THE FACT OF MORAL EVIL

135

to become a spy. And God knows what some being A which he did not create would have done if he had been created
instead of me, with my genes and brought up in my family circumstances. God's present knowledge of how beings
which he did not or will not create would have chosen if he had created them, and of how beings which he did or will
create would have chosen in circumstances which he will not bring about, is called knowledge of counterfactuals of
freedom. These are conditional assertions of the form Tf p then tf, where p> describes circumstances which, it implies,
did not and will not occur, and ‘tf describes an action which some agent would do freely, either some actual agent or a
possible agent.

But, if God knows in advance how all possible beings would choose in all possible circumstances, then, since it might
seem as if there were infinitely many such beings and infinitely many such circumstances, surely God, who does not
want people to choose what is wrong (and only gives them choice in order to give them the opportunity of choosing
rightly), would only create beings and put them in circumstances in which (he foresaw) they would freely choose the
good. To save the free-will defence against this objection, Plantinga draws our attention to a further possibility:
‘universal transworld depravity*. Maybe, he suggests, each of the (presumably infinite number of) possible beings
whom God could create would in some of the circumstances in any world in which he created them freely choose what
is wrong . 110 It is logically possible, Plantinga urges, that there is universal transworld depravity. He does not claim that it
is probable that there is such depravity, but merely that, because this is possible, it does not follow from God having
middle knowledge that he can create a world in which there are free beings who always choose the good. And that
seems correct.

But, as R. M. Adams among others has suggested, the hypothesis of a universal transworld depravity looks highly
implausible . 111 Surely there are possible beings who could live in some circumstances so that throughout their lives they
freely chose only the good. Although no one, not even God, could cause that to happen, he could, if he had middle
knowledge, create only those whom he knew would so choose. If you were allowed to create an infinite

110 The Nature of Necessity, 1 84-9 .

111 See R. M. Adams, ‘Middle Knowledge and the Problem of Evil’, American Philosophical Quarterly, 14 (1977), 109—17: 116-17 .

136

THE NECESSARY EVILS

number of free agents and put them in any of an infinite number of different circumstances, and then see how they
would act, surely you would find a few agents who in some circumstances would do nothing but good. But if God
could see in advance, before creating agents, how they would act in different circumstances, he could create just those
few and put them in just those circumstances where they would freely choose the good. The infinity of the number of
beings and possible choices available to each in lives in different worlds shows that if this logical possibility of
transworld depravity is to constitute a serious counter-objection to the difficulty raised for the free-will defence by the
suggestion that God has middle knowledge, then the onus is on Plantinga to prove, or show it to some extent
probable, that all creatures would do some wrong in some circumstances. Even if all creatures will do wrong some of
the time, could not God give them freedom only at such times at which, he foresees, they will freely choose what is
right? Plantinga would have to respond by saying that possible creaturely decisions are so interwoven — necessarily, so
that ‘the more God withholds freedom so as to prevent evil, the more evil would be done on the remaining occasions
when creatures are left free’. But as David Lewis, who offers Plantinga this way out, comments, that is, though perhaps
possible, ‘not plausible, except as a last resort for heroic faith ’." 2 It would seem to follow that if God did have middle
knowledge, he could give humans on many occasions libertarian free will without their doing wrong; and he could do
this knowing that no wrong would result. So the possibility of moral evil would not be required if God is to give to
creatures libertarian free will . 113

Adams has suggested that there are no counterfactuals of freedom, or at any rate ones with truth-value (i.e. ones which
are true or false). In that case not even an omniscient being could know which counterfactuals of freedom were true or
false and thus act to put creatures only in those circumstances in which they will freely choose righdy. Plantinga has
urged , 114 however, that there must be counterfactuals of freedom with truth-value, for often we can have

112 David Lewis, ‘Evil for Freedom's Sake’, Philosophical Papers, 22 (1993), 149—72: 162 .

113 Creatures might still need the natural evil of bad desires, which they resist, if they are freely to choose the good instead of the bad.

114 See J. Tomberlin and P. van Inwagen (eds.), Alvin Plantinga (D. Reidel, 1985), 372-82 .

THE FACT OF MORAL EVIL

137

a justified belief about what is the truth-value of some counterfactual of freedom. Knowing the strength of the desires
to which some agent would be subject in certain circumstances, we may often justifiably predict that in those
circumstances he would yield to them, even though he was free not to yield and could, if he had tried hard, have
resisted the desires. Against Plantinga and in support of Adams, William Hasker 115 raised the question, if a
counterfactual of freedom has a truth-value, what makes it true?

Consider the counterfactual ‘If Elizabeth were offered the grant, she would accept it’. Suppose the counterfactual to be
true. Being a counterfactual, this presupposes that Elizabeth is not offered the grant. So what makes it the case that the
counterfactual is true? If the world were deterministic, its laws of nature would make it the case that there would be a
certain outcome if Elizabeth were offered the grant; and it would be the fact that they operated which would provide
the truth-conditions of the counterfactual. But if it is a counterfactual ‘of freedom’, there are no such laws. Yet
whatever, if anything, does make the counterfactual true, it will not be Elizabeth, for (ex hypothesi ) she has not been in a
position to make the choice. But, Hasker urges, if it is not a counterfactual made true by Elizabeth's free choice, it
cannot be a counterfactual ‘of freedom’ at all, for she is the agent with whose freedom it was supposedly concerned. If
we suppose the counterfactual false, we reach the same conclusion that it cannot be Elizabeth's free choice that makes
it false. However, the defender of counterfactuals of freedom may urge that nothing else makes the counterfactual
true; it is just ‘barely true’. Take a different sort of example. Suppose that it is not predetermined whether tossed coins
land heads or tails. I did not toss a coin yesterday; but perhaps there is a truth about how some coin would have landed
if I had tossed it.

So let us grant this point, and suppose that there are counterfactuals of freedom with truth-value. Even so, Plantinga
has gone wrong in supposing that anyone could have incorrigible knowledge of the truth-value of a counterfactual of
freedom. And this for the reason that no one can foreknow incorrigibly the future free actions of an agent in
circumstances which will be realized. My argument for that is not original , 116 but it is, to my mind, of irrefutable

115 See his God, Time and Knowledge (Cornell University Press, 1989), 39-52 .

116 This argument was set out very carefully and cogently by Nelson Pike, ‘Divine Omniscience and Voluntary Action’, Philosophical Review, 74 (1965), 27-46 .

138

THE NECESSARY EVILS

cogency. Consider such a free agent / faced with a choice tomorrow of doing x or not doing x; and any pre-cognizer G.
If G were to know today what / will do tomorrow, he will have a belief about this — for knowledge entails belief. Given
that, of logical necessity, the past is unaffectable by present actions (causes cannot follow their effects), 117 then G's belief
today will be what it is before and independently of what J does tomorrow. Yet what J does tomorrow depends on f s
choice tomorrow. If / is indeed free, he is free to make G's belief, whatever it is, false. He may not do so, but he has it
in his power to do so, and so G's belief cannot be necessarily true, and so cannot amount to incorrigible knowledge. If
no one can have incorrigible knowledge of the future free actions of actual agents, a fortiori they cannot know what they
will do in unrealized circumstances, let alone what possible unactualized agents will do. If no one can do all this, even
God cannot.

Plantinga has subsequently claimed 118 that God's belief today about what / will do tomorrow is not what it is before
and independently of what J does tomorrow. More generally, God's beliefs at a time t are not ‘hard facts’ about t. A
‘hard fact’ about a certain time is a fact, all of whose truth-conditions, the states of affairs which make it a fact, are
states of affairs at that time. A ‘soft fact’ about a certain time, by contrast, is a fact whose truth-conditions are (at least
in part) states of affairs at some other time. 119 Thus ‘John played the piano in 1984’ (if true) expresses a hard fact about
the world in 1984; whereas ‘The next century's greatest pianist played the piano in 1984’ (if true) expresses a soft fact
about the world in 1984, for its truth depends partly on how things go in the next century. The Times having a
prediction today about who will win tomorrow's Derby is a hard fact about today. But The Times having the correct
prediction today about who will win tomorrow's Derby is not a hard fact about today, because whether it has a correct
prediction today depends partly on (i.e. is partly constituted

For argument on this, see my The Christian God (Clarendon Press, 1994), 81—90 .

A. Plantinga, ‘On Ockham’s Way Out’, Faith and Philosophy, 3 (1986), 235—69 .

My definitions of ‘hard’ and ‘soft 5 facts, taken from The Christian God , 132, are somewhat different from those of most writers, who construe a ‘hard fact’ about a certain
time as one all of whose truth-conditions are states of affairs at or earlier than that time, and a soft fact as one all of whose truth-conditions are states of affairs later than that
time. This difference does not affect the present argument.

THE FACT OF MORAL EVIL

139

by) what happens tomorrow. So, Plantinga suggests, while my having a certain belief today about what will happen
tomorrow is a hard fact about today, God s having a certain belief today about what will happen tomorrow is not: it is a
soft fact, because God's beliefs today about tomorrow are the beliefs they are in virtue of what happens tomorrow.
The claim of the last paragraph that causes cannot follow their effects is to be read as the claim that whether some hard
fact holds cannot be affected by what happens later; soft facts can be so affected. Hence, Plantinga claims, the
argument of the last paragraph is unsound, and God can have an incorrigibly true belief today about what I will do
freely tomorrow. 120 So, Plantinga puts himself in a position again to defend the logical possibility of God's having
middle knowledge, and so to need the hypothesis of transworld depravity in order to explain the occurrence of moral
evil.

However, the suggestion that God having a certain belief today about which action / will do tomorrow is a soft fact
about today seems to empty of all content the notion that God has a ‘belief’ today. Of course whether the belief is true
depends on what happens tomorrow, but what the belief is does not. That being so, it cannot be that God has
necessarily true beliefs today about what / will do freely tomorrow; and more generally God cannot have incorrigible
knowledge of the future actions of free creatures, let alone the actions which they would perform in circumstances
which will not occur, or the actions which creatures never to be created would have performed if they had been
created. So if God gives us freedom to choose between right and wrong, he cannot foreknow incorrigibly how we will
choose; and so he inevitably allows the possibility of our doing wrong. (And if the freedom he gives us is limited to a
choice between good and the bad which is not wrong, he inevitably allows the possibility of our choosing the latter.)
And if he gives freedom for much time to many creatures, he makes very probable the occurrence of much moral evil.
Plantinga's affirmation of the latter stands secure against the difficulties which he raised for his own position. Since
Christian tradition has on the whole affirmed that humans have free will (see Chapter 2) God's omniscience must, I
suggest, be understood (as I suggested in Chapter 1) as knowledge at every time of all that is

See Additional Note 4 .

140

THE NECESSARY EVILS

logically possible for God to know at that time. This knowledge does not include knowledge of the true propositions
about the future actions of free agents. 121

Temptation

We saw in Chapter 5 that for agents 122 to have free will of any kind other than very unserious free will, they need desires
to do actions of strengths out of proportion to the goodness of the actions desired. To have a free choice between a
greater and a lesser good, they need a desire for the lesser good stronger than that for the greater good. To have
desires out of proportion to the worth of the action desired is in itself a bad thing — mildly so. But to have a free choice
between good and bad, agents need a desire for the bad stronger than for the good. Desires for the bad are bad. And
to have a free choice between good and wrong — what I call very serious free will — agents need a desire for what is
wrong stronger than for the good. The wider the range of free choice — in itself so good a thing — the worse the desires
required to make it possible. It is bad to want to deceive, rape, hurt, maim; there is something badly amiss with an
agent whose attitude for his fellows is to seek their harm in these ways, an attitude to which he will inevitably yield
unless he makes the free effort to hold such urges in check. And the stronger the bad desires, the worse is the agent's
condition.

In my view an action would not be intentional at all unless done because it was regarded as in some way a good thing.
Intentional action is action seeking to bring about an effect which the agent aims at because he regards its fulfilment as
good to achieve. If I believe that going to London would give me no pleasure, would give no one else any pleasure,
would not enable me to do something else which was in some way good, would not enable me to avoid doing
something in some way bad, would not involve fulfilment of duty, etc., then the movement of my limbs in the direction
of London cannot be regarded as an intentional action. It would follow that we cannot aim at the bad as such, but what
we can do

See Additional Note 5 .

Other than an agent such as God in the special condition described in Ch. 5

THE FACT OF MORAL EVIL

141

is to aim at some good which can only be achieved by means of something bad, and so aim at what is bad, despite its
badness. He who aims to deceive seeks some good (normally for himself) which the deception will achieve; he who
aims to rape seeks thereby the good of sensory sexual pleasure; and even he who aims to hurt aims thereby at the good
of punishment for wrongdoing or the exercise of his own power (a good in itself). The trouble is that these
actions have effects or concomitants which mean that in pursuing a good agents aim at what they truly believe to be
overall bad.

Although I do believe this — that all intentional action is action to realize a believed good — my present argument does
not depend on it. It depends solely on the view advocated in Chapter 5, that if we do believe some action is overall the
best, we will do it, unless non-rational forces, i.e. desires, provide a counter-influence. For humans to have libertarian
free choice between good and bad, not merely is the possibility of moral evil required, but the actual occurrence of a
certain kind of natural evil — bad desires — is required.

Bad desires, and above all wrong desires (as well as good desires), are an evident human phenomenon. They are often
(though not always) desires for kinds of things which are good to possess in some circumstances but not in all, but the
desires operate both when it is good that we indulge them and when it is bad that we indulge them. Often it is good
that we should have and so desire to have food and drink (when we need them), sexual intercourse (with our spouse),
fame (when we deserve it), and fortune (when we can rightly use it). The trouble is that we have the desires when it
would be bad, and often wrong, to indulge them. And yet this natural evil is a necessary condition of the good of free
choice between good and bad. I cannot exercise a serious free choice to give money to the starving unless I have some
desire to possess money when it would be good to give it away (i.e. a certain miserliness). Since it is so often good to
act, and less often good to do nothing, that inclination which will so often ensure that we have a free choice between
good and bad is sloth, in the modern sense of idleness or the desire to do nothing. (I shall have more to say about the
importance of sloth in the next chapter.)

The temptation to yield to a bad desire will be greater if the desire is the desire to do some action immediately. Since
desires by their very nature act upon us, whereas we have to take positive action to resist them, it will be easier to resist
bad desires if we have

142

THE NECESSARY EVILS

time to pull ourselves together. We are more likely to yield to desires which lead more instinctively to action.

I have been assuming (see Chapter 5) that we know the consequences of our actions and their moral worth. But of
course we may not; and in that case the serious free choice is between the believed good and bad. Our situation is,
however, very often an intermediate one. We do not know the consequences of our action, nor are we entirely ignorant
of them. We believe it slightly probable or fairly probable that they will have such-and-such consequences; and there is
quite a chance that these are bad consequences, and indeed that it is wrong for us to bring them about. The good
desires are weaker and the bad ones stronger in so far as we doubt that the bad action will have bad effects. In a
situation of doubt about consequences, the temptation also arises for self-deception, for refusing to admit to ourselves
that the probability of the bad consequences is not really as high as in our initial honest moment we believed that it
was. If the evidence only suggests that probably smoking causes cancer, we can tell ourselves that nothing definite is
known about the matter. Uncertainty clouds, and allows us to cloud even more, the stark choice between exposing
ourselves to a high risk of cancer and not smoking.

The choice between what is probably good and what is probably wrong is a very serious choice, but a blurred one; and
does not provide the opportunity for going so deliberately against the good as does a choice where we know clearly
what we are doing. Both the choice being a considered one and the wrongness of a wrong action being obvious are
two marks of the choice of wrong being a mortal as opposed to a venial sin in the traditional distinction. The other
mark is that the mortal sin involves worse wrong than the venial sin. The more the choice between right and wrong is
considered and evident, in a way the better. Our libertarian free will is freer, more obviously in control of the way the
world is to go. But the worse it is if we make the wrong choice; for we have set ourselves more deliberately against the
good. It is good to be spared such stark choices in the early years of childhood; but good too to have the choice of
improving our self-control (so as not to be swept away by desires) and our moral and factual knowledge, so as to make
stark choices which express our stand.

I argued in Chapter 5 that the goodness of doing the best action is better, the stronger the contrary temptation and the
more significant

THE FACT OF MORAL EVIL

143

the good act, without espousing any view about how strong a temptation and how important the alternatives it would
be best for a creator to give creatures, in view of the risk of bad resulting — both the bad intentional acts and the bad
consequences following therefrom. There follows not merely the formal logical impossibility of God giving us free will
to choose between good and bad without bringing about the natural possibility that (unprevented by God) we will do
bad, but a whole range of more detailed logical impossibilities. God cannot give us certain kinds of free will (certain
strengths of temptations to choose between certain kinds of important actions) and at the same time ensure that there
is only such-and-such a probability that we will do such-and-such bad or wrong actions. The stronger the temptation
to do bad, and the more significant are the good or bad actions, the greater the possibilities for good that God gives us
and the less the chance that those possibilities will be realized.

This chapter has been concerned with intentional action as such, and in particular with free intentional action; and with
the bad (and especially the wrong) inevitably involved in various kinds of good free choice, differing in seriousness and
strength. The goodness of free intentional action will vary also, as we have seen, with its significance, with the kinds of
difference which it makes to the world, the responsibility for which belongs to the agent in his choice. The next chapter
will consider the various kinds of bad involved in various kinds of responsibility.

8 The Range of Moral Evil; and Responsibility

See, I have set before thee this day life and good, and death and evil; in that I command thee this day to love the
Lord thy God, to walk in his ways, and to keep his commandments and his statutes and his judgements, that
thou mayest live and multiply, and that the Lord thy God may bless thee in the land whither thou goest in to
possess it. But if thine heart turn away, and thou wilt not hear, but shalt be drawn away, and worship other gods,
and serve them; I denounce unto you this day, that ye shall surely perish; ye shall not prolong your days upon the
land, whither thou passest over Jordan to go in to possess it. I call heaven and earth to witness against you this
day, that I have set before thee life and death, the blessing and the curse: therefore choose life, that thou mayest
live, thou and thy seed: to love the Lord thy God, to obey his voice, and to cleave unto him: for he is thy life, and
the length of thy days: that thou mayest dwell in the land which the Lord sware unto thy fathers, to Abraham, to
Isaac, and to Jacob, to give them.

(Deuteronomy 30: 15-20)

The Principle of Honesty

So the good of many agents having many free choices between good and bad involves the very high probability of
much moral evil — the bad choices unprevented by God. But God could of course arrange things so that our bad
choices never had any effects. When we chose kind words, they came out of our mouth; when we chose to insult, the
air did not convey the message. When we chose to strike, we became paralysed; when we chose to stroke, our hands
obeyed our commands. But for God to create agents permanendy so placed

THE RANGE OF MORAL EVIL

145

would be a great deceit. He would have made it seem to us as though we had power — for we could not have a choice
between uttering kind words and uttering insulting words, unless we thought our attempts to talk would be successful.
God would not be the parent (of p. 93) who warned the elder son that he would intervene the moment the son did a
thing wrong, but the parent who always gave the impression that he would not intervene, but always did the moment
the son made a bad choice.

It is, I suggest, a moral truth which I shall call the Principle of Honesty:

God has an obligation not to make a world in which agents are systematically deceived on important matters

without their having the possibility of discovering their deception.

The principle is not concerned with what it is right for God to tell people, e.g. in words, by a propositional revelation
(my view on that is that it would never be right for him to tell us anything false); but with what it is right for him to
allow people to infer from the world which he makes. It does not rule out him allowing one person to get a false idea
of how the world works; or allowing one person to tell another falsities about how the world works. But it does rule
out God making a world which leads people to have false views on important matters, which they cannot discover to
be false.

The principle is a natural extrapolation to God from the similar obligations to honesty of a human teacher. It would
always be wrong for me to give a student a false piece of information, which I state to be true. But it is not always
wrong for a teacher to include misleading elements in what he shows to his students, in order to force them to sort
things out for themselves. I may give a student an apparently valid but really invalid argument, without saying either
that it is valid or that it is invalid, in the hope that he will come to discover for himself that it is invalid. But it would be
wrong for me to allow a student to get a false impression from my teaching if there was no possibility of the student
finding out the truth. It would be wrong for a teacher to present a student with an apparently plausible account of
history or geography, in which the various strands fitted with each other in such a way that the student could never find
out that it was basically false. If the teacher is teaching a class, rather than individuals on their own, the

146

THE NECESSARY EVILS

possibility of discovery will be a class one: to discover the falsity may require cooperation between members of the
class.

Now while individual humans teach other humans (or groups thereof) over a short time-scale, God deals with the
human race as a whole over a long time-scale. Although he may sometimes intervene in the world, in general he has
created a world where things behave in regular ways; and it is individual humans who try to find out how things behave
and teach others, and who try to discover moral truths and teach others. And of course the others may come to
correct what they have been taught. We have already commented on the goodness of such an arrangement in Chapter
5. Therefore what the Principle of Honesty says is that God must not create a world in which in general people cannot
discover the truth of how it works and what is good and bad, at any rate over time, in cooperation. It allows the
possibility of error (factual and moral) which some can help others to discover — though not, I would suggest, too
much even of that.

For God to make a world in which we think that some of our actions cause suffering when really they do not, or do
not cause suffering when really they do, when the human race could never discover differently, would be to treat
creatures with contumely. Likewise if God is to give us a choice between actions which seem to matter, say by causing
pain or pleasure, they must in general matter — namely, cause pain or pleasure; or we must be able to find out that they
do not. Just how soon we must be able to find out is a difficult matter (as are all matters of drawing exact borderlines
for moral principles); though I would suggest that most, though not all, errors ought to be discoverable by individuals
or communities within a few years. But God must not put us in a world of virtual reality which we could never
discover to be such. The Principle of Honesty is a moral truth very important for the concern of this book, and we
shall meet it several times in the next few chapters. It limits the ways in which God can give us good without bad. 123 Its
immediate consequences at this stage of the argument are that God cannot give us apparently serious moral choices of
doing good or bad and yet ensure that our choices never make any real difference.

: would be wrong of God to deceive us is a
' ‘systematically’.

That it

point central to Desc

philosophy (see R. Descartes, Meditations 4, 5, and 6); but he does not add the qualifying

THE RANGE OF MORAL EVIL

147

The Bad Involved in Responsibility for Oneself and the Consequen-
ces of Ignorance

But, we saw in Chapter 5, the more significant the choices an agent has, the better it is for him. And indeed, even if it
were permissible for God to create a deceptive world, it would be better for us if our choices do really make significant
differences rather than our merely thinking that they do. So it is good for me if I can influence my present condition
and my future condition — really influence it, and that means for good or ill. For God would have given me no real
responsibility for myself if he ensured that I could do myself no harm. God will allow me to sustain my strength by
eating or to neglect to do so. And God will allow me to do actions which risk harming myself, as well as ones which
will certainly do so: to go rock-climbing or consume dangerous drugs. And God will allow me to do actions which will
have (or probably have) long-term effects on myself for good or ill. He will allow me to ruin my health by smoking or
alcohol, or to care for it with proper diet and exercise.

But if our choices are to be choices which make a difference to things for good or ill — not just choices made in a
simulator — we need, as has been pointed out earlier, knowledge, factual and moral, of the consequences of our
choices. And, as we saw in Chapter 7, our freedom is greater if we really realize what we are up to: we do not yield to
the desire to do wrong through a cloudy perception of its wrongness. So an important choice which we each of us have
with respect to our own future is the choice of whether or not to try to acquire knowledge, factual and moral. If we are
to have these choices, we must begin in a position of (at any rate) relative ignorance. I cannot choose whether or not to
bother to find out the effects of smoking if I know already. And I need ignorance of what is morally good and bad if I
am to choose whether to seek the knowledge. Though in general ignorance is simply the lack of a good state, not the
presence of a bad state, it is an absence of a good state necessary for choice. But certain sorts of ignorance will make it
immensely more likely that I will act or neglect to act, with effects that I do not desire. If I do not know the effects of
smoking, then I am much more likely to get lung cancer, something I do not desire. And ignorance of that kind would
seem a bad state, not merely the

148

THE NECESSARY EVILS

absence of a good. And moral ignorance with respect to actions which I am likely to perform would also seem a bad
state. It is bad if I do not think that it matters whether or not I hurt others or indeed myself. Yet if the great choice of
whether to try to acquire knowledge, factual and moral, is to be open to me, I may not bother; and the bad
consequences will flow.

I shall have much to say about factual knowledge and the way open to us to improve it in Chapter 10. The possibilities
through experience and discussion for acquiring moral knowledge surround us, and thus we have the opportunity to
follow them up or to hide from ourselves moral truths which they seem to be indicating. We can, through honest
thought, cooperative discussion, and hard experience, come to have a deeper and deeper understanding of values; or,
through idleness and self-interest, allow ourselves to have a trivial and false understanding of values, and even come to
believe that there is no right and wrong. This good, of it being up to us whether we acquire a morality and of what
kind, involves the necessary evil to begin with of us having inadequate moral views, and those views leading to bad
actions, for example to cruelty in war against enemies regarded as entitled to no human rights because they are
foreigners.

Growth in morality involves first understanding justice, that there is right and wrong; that we have duties to repay our
debts and not kill or maim our fellows. And it involves understanding that if we do kill or maim, that gives others the
right to take back from us an equivalent of what we have taken away — ‘an eye for an eye, and a tooth for a tooth’.
Justice involves the right to punish. 124 But further growth in morality involves understanding that it is sometimes good
to let people off their debts and not exercise our right to punish them — good, that is, to show mercy — and generally to
be generous to others beyond the call of obligation. We could not understand the latter without understanding the
former: that is a logical point. We could be born with both items of knowledge; or we could help each other (through
discussion and encouragement) to move from the former narrower understanding to the wider one (or refuse to do
so). Only the latter method makes moral outlook a matter for choice. So there is a point in a society starting with the

124 For fuller discussion of this claim about the sources o
(Clarendon Press, 1989), chs. 5 and 6 .

}f the right to punish, its extent, and the frequent goodness of \

ly Responsibility and Atonement

THE RANGE OF MORAL EVIL

149

morality of the Books of Exodus and Leviticus and then gradually being encouraged to adopt the teaching of the
major prophets and then to develop the morality of the Sermon on the Mount.

It is good that we should have power over ourselves, gradually to mould our desires for good or ill. But if we have this
power, then there is the danger that we will let ourselves acquire bad desires, with all the ill that that constitutes and
which will follow from it. The greater our power to affect our desires, and the greater the range of desires — good and
bad — open to us, the better, in virtue of the choice that it makes available to us. But of course the more such power we
have, the greater the probability that some of us will allow ourselves to have bad desires and indulge them: racists and
misogynists are likely to emerge. It is good, as we saw in Chapter 5, that we should have the power (as apparently we
do) gradually, over time, to mould our characters so as to fix them permanently. Our characters consist of our beliefs,
principally our moral beliefs, and our desires. We saw there how, by always choosing the good, we can so weaken the
force of bad desires that they cease to have influence over us. Now God could have given us merely the choice of
whether to form a good character or to postpone doing so, so that the choice always remained open in future to
initiate that process. But a God who only gave us that choice would have kept back from us the further possible
choice — of so immunizing ourselves to any moral influence that we no longer had the possibility of forming a good
character. A God who did not give us the latter alternative would have rigged the choices open to us in his own favour,
i.e. in favour of the good. We saw in Chapter 5 that it does rather look as if humans have open to them this final
alternative, of always so indulging our strongest desires (including a considerable number of bad desires) that we lose
the power to choose between alternatives in virtue of their perceived goodness. Instead, we become a theatre of
competing desires, often bad desires, in which the strongest always wins. It does rather look as if the evil men and
women of history have got themselves into this condition. 125

What will happen to our moral beliefs in this situation? Will we still believe that there are good actions to be done, but
regretfully acknowledge that we have lost the power to choose them? Or will we cease to believe that there are good
actions to be done? Would

125 See Additional Note 6 .

150

THE NECESSARY EVILS

the failure to obey conscience leave it nagging at us for ever ineffectually, or will it lead to its elimination? Either
alternative seems logically possible. Empirical psychology suggests that what in fact happens on a small scale in all of
us when we continually refuse to do what we believe to be obligatory or even good in some respect is that we find the
situation of mental conflict between belief and action — between our beliefs about what is good to do and what we in
fact do — intolerable. So, unless we in fact sometimes do the good, we hide from ourselves its goodness . 126 Suppose,
being white, I find myself with a desire to abuse, insult, and subjugate black people; but also the belief that this is a bad
desire to have. The more I indulge my desire, the more indulging it becomes natural, and then the less I am willing to
admit to myself that it is a bad desire to have. That is compatible with my still believing that it is a bad desire to have
while refusing to admit it to myself, but it is also compatible with my having eliminated the desire. I do not think that
we have the empirical evidence about which is the final state of bad humans who have so allowed their character to be
corrupted that they have lost the power to choose what is good because it is good.

Which of these alternatives would a good God produce in those who have chosen a bad character? It is good, we have
seen, to desire the morally good, and bad indeed to have no moral sensitivity. It is also bad to be unable freely to
choose what we recognize as morally good, for our conscience to be thus incapacitated. Clearly the latter state is the
worse; the former state is simply the absence of a good. It may be, however, that there is something very bad in God
allowing us to lose something so valuable as our moral sensitivity, but on balance I would think that a good God in
allowing us to become a prisoner of our desires would allow us to eliminate our awareness of goodness, perhaps not
ever on Earth, but in the end. The medievals disputed whether it is possible for our awareness of moral goodness to be
eliminated. Aquinas argued that it is not possible, but, as I interpret him, Philip the Chancellor argued that it is
possible; and in this dispute I side with the latter . 127 Either way, the great moral evil of an evil character may very well
occur if we have the great

126 For fuller philosophical and empirical justification of this claim about the possibilities of corruption open to humans, see my "Responsibility and Atonement , 173-8.

127 See the extracts from Philip the Chancellor and Aquinas in T. C. Potts, Conscience in Medieval Philosophy (Cambridge University Press, 1980), 108-9 and 129-30 .

THE RANGE OF MORAL EVIL

151

good of the free choice of what character to have. But of course God would only be good in giving us this terrible
choice, if he allows us some long time in which we have many (but not endless) opportunities of turning back; he must
thereby ensure that our commitment to it is firm. 128 Christian doctrine has normally held that we do have the
opportunity so to immunize ourselves that we are no longer able to choose to seek the good of the Beatific Vision in
the Heaven of the Blessed.

Given my earlier claim that a good act is better if done contrary to temptation, it might seem open to question whether
it is a good thing to seek to encourage good desires and discourage bad desires, for that — if successful — would remove
the possibility of resisting temptation. However, we do not have the right to put ourselves in situations where we are
likely to do wrong, and so we do not have the right to seek a temptation to do what is wrong. If the temptation to beat
my wife becomes very strong after I have drunk two glasses of whisky, I do not have the right to drink two glasses of
whisky. Obligations must be fulfilled first before we seek any greater good. 129 ‘Lead us not into temptation’ is indeed
the proper Christian prayer. God, however, has the right to cause us to be tempted to do those actions (which include
harming others) which God has the right to allow us to do. For if it is right for A (parents or the state, say) to allow B
to do some action which B ought not to do, it must be right for A to allow B to be tempted to do that action, for
unless B is tempted to do a wrong action, he will not do it, and so preventing temptation is preventing the action. And
if God has the right to cause us to be tempted, a fortiori he has the right to allow others to cause us to be tempted. The
extent of God's right to allow others to do wrong actions will be considered in Chapter 12. Yet, even though it is not
right to seek temptation, it may well be appropriate to rejoice when it is foisted upon us, because of the opportunity
presented to us to resist it. St James tells his readers to rejoice ‘when ye fall into manifold temptations’. 130

128 This provides the answer to Scobie's concern: ‘It seemed to Scobie that life was immeasurably long. Couldn't the test of man have been carried out in fewer years? Couldn't
we have committed our first major sin at seven, have ruined ourselves for love or hate at ten, have clutched at redemption on a fifteen-year-old death bed?’ (Graham
Greene, The Heart of the Matter (Heinemann, 1957), 56) .

129 And if there is a God, the scope of obligation is much larger than it would otherw

130 Jas. 1: 2.

vise be. (See my Responsibility <

, ch. 8.)

152

THE NECESSARY EVILS

Where the temptation is only a temptation to do something less good, or bad but not wrong, it could in theory on
occasion be good to seek to be tempted and so to allow bad desires to continue, or even to develop them. But the
probability of achieving the good of overcoming temptation would need to be weighed against the intrinsic badness of
the bad desires, the probability of yielding to the temptation with the resulting bad consequences (two bad states),
and — since one bad act today makes it easier to do a bad act next time — the increasing probability of forming a bad
character. The circumstances would have to be very special indeed for it ever to be good to seek to be tempted even to
do what is bad but not wrong . 131 Only under the most exceptional circumstances would a good agent not seek to
eliminate their bad desires. A good God, seeking for agents choices of great significance, will allow them the choice of
whether or not to do so.

One of the greatest of all our responsibilities in respect of ourselves is the responsibility for our own life or death. We
can through negligence shorten our lives, by drinking ourselves to death, for example; or through taking care, extend
them. And we have, almost all of us, the power to end life there and then — to commit suicide. I hope that my readers
will forgive my dwelling for a little here and subsequently on this grim choice which is ever available; for it seems to me
a very important aspect of the human condition that it is available, and its availability is an important strand in the
solution to the problem of evil.

The choice between life and immediate death is one of the most significant of all the choices open to us. God has not
made humans naturally immortal. We can end each other's lives. For most of us it is quite easy to end our own lives, if
we so choose: by jumping off

I have argued elsewhere {Responsibility and Atonement, 158—9 ; The Christian God (Clarendon Press, 1994), 203—7 ) that God, in becoming incarnate in Jesus Christ, took upon
himself desires to do lesser rather than supererogatory goods. He did this in order that, by overcoming the temptations to do only the lesser goods, he might atone for our
sins. However, I sometimes expressed the point misleadingly by writing that he acted with the foreseen consequence that he became subject to the desires for lesser goods; I
should always have written that he intentionally subjected himself to those desires. Such an action was indeed one done under the most exceptional circumstances. Maybe
too the prayer of St Ephrem (as recorded by St Peter of Damascus, in G. E. H. Palmer, P. Sherrard, and K. Ware (ed. and trans.), Philokalia, iii (Faber & Faber, 1984), 152)
to be given temptations in order to overcome them was a prayer which concerned temptations to do lesser goods, made under exceptional circumstances! But we have no
reason to suppose this.

THE RANGE OF MORAL EVIL

153

a tall building or throwing ourselves beneath a railway train. These may be, we reasonably judge, painful ways of
ending life. And for some few, the disabled, the only ways of ending life unaided are the processes of starvation which
take weeks. There are (in the painfulness of the process) these and other deterrents to suicide, and we will consider
them later in the chapter. And if we have any relatives or friends who love us or are dependent on us, suicide is a very
wrong act. And if there is a God who gives us life, it is always a wrong, for it is throwing back a very good gift in the
face of the giver. But if there is a God, he has made available to us the possibility of this wrong act; and it is a mark of
his generosity that he has done so. It is, we shall see in a later chapter, part of what gives him the right to allow us to
suffer for the sake of a greater good.

The Bad Involved in Responsibility for Others

But what an awful world it would be if the only good or harm we could do was to ourselves. It is of course a good for
us to be able to mould the inanimate world, but clearly a much greater good is to have responsibility for other animate
beings — and that serious responsibility involves again the ability to benefit or harm them, as we saw in Chapter 5. If
my responsibility for you is limited to whether or not to give you some quite unexpected new piece of photographic
equipment, but I cannot make you unhappy, stunt your growth, or limit your education, then I do not have a great deal
of responsibility for you. For the way things go with you will not in that case be largely up to me. Other factors will
determine how it is with you; I will only be able to tinker a bit, and top up your well-being. If I am to have significant
responsibility for you, I have to have the alternative of hurting and maiming as well as the alternative of greatly
benefiting. But if free agents are to have this responsibility, there is a high probability that others will be harmed, and
that often means suffering. The more free individuals there are and the more responsibility they have, the greater the
probability of more harm to many.

We cannot do good or harm to each other in many ways without humans already cooperating in the acquisition of
knowledge. I cannot discover by myself many of the consequences of my own

154

THE NECESSARY EVILS

individual actions. The world is so arranged that much cooperation is needed to discover the long-term consequences
of smoking etc. And it is good, for all the reasons discussed in Chapter 5, that things should be thus. What I can
choose to do or not to do is to choose whether to tap into the process, and acquire the knowledge which others have
produced; and also to choose whether or not to help in the process of knowledge acquisition. And so many goods and
harms which we know how to bring about can only be brought about by the cooperation of many humans, sometimes
over many decades.

It is good that humans have a responsibility for animals. A child who is given a hamster to care for is benefited, when it
all depends on the child whether the hamster is fed and watered at night or not. The child is pleased to be trusted by
parents with the hamster; and by an argument of a kind which I have used a number of times, if the child is right to be
pleased, that can only be because it is a good thing that the child is trusted. More generally it is good for humans to
have the responsibility for the animal kingdom. And it is good for more mature humans to have the deeper
responsibility of choosing whether or not to produce children, and whether and how to educate and nourish them.
Again, if there is always someone else in the background who will step in to save those children who are neglected and
abused, parents will not have the serious responsibility of care. The good of serious responsibility makes almost
inevitable many great moral evils.

To prevent misunderstanding at this stage, I should make clear again that I am not saying that a good God would give
creatures unlimited responsibility for others; only that there is a trade-off: the more responsibility for some, the more
probable harm to others; the less probable harm, the less responsibility. Nor am I saying that responsibility for children
ought to rest solely with parents. There may or may not be a good case for a strong team of social workers keeping an
eye on ‘problem families’. The responsibility for children would in such a case still remain with humans — but with a
wider group of humans being responsible for each child. Yet interference by social workers, just like interference by
God, does diminish the responsibility of individual parents.

As we saw in Chapter 5, it is good that the web of responsibility should extend to parents, relations, and friends; and,
further, that we should be able to make a difference to people in distant places

THE RANGE OF MORAL EVIL

155

and future times. But once again, inevitably of logical necessity, the more responsibility free agents have, the greater the
probability of much moral evil. We can allow bad states to occur either by doing an action which, we foresee, will bring
them about; or by neglecting to do an action which, we foresee, would stop them occurring. So many of the world's ills
(e.g. recent mass starvation in Africa, foreseen by relief agencies who warned governments) seem to be in the latter
category. And as human control grows, so those and more bad states which in earlier ages would have been natural
evils move into the category of moral evils, because preventable by the actions, if not of those now living, of earlier
generations.

We are born into a world in which the good things of life are divided unevenly — by our genes, initially. Some are born
so as naturally to develop into physically strong, clever, or beautiful people; others, not so. And society so often adds to
the unevenness of the distribution of the good. The physically strong, clever, and beautiful naturally (unless prevented)
acquire enormous power over others. The lame need help, but the beautiful command with a smile. There is no reason
why a good God should distribute the good things of life evenly. If people contribute in some sense equally to a
common enterprise, there is something wrong if they do not receive an equal share in its benefits. But if our creator
creates us out of nothing, and gives to some ten good things, and to others twenty good things, no one is wronged; nor
has he failed to be perfectly good. He has been generous, and, more so, he has made it possible for us to be generous.
For the more power someone has and the more someone else lacks it, the greater the opportunity the former has to
use his power to benefit the latter and above all to increase the power of the latter. If you are strong, you can use your
strength to do things for me which I cannot do for myself. If you are beautiful, you can use your beauty to make me
happy. If you are clever, you can use your cleverness to devise ways of making me clever. If we all had the same
number of the same good things, there would be so much less scope for helping each other.

It is good, we saw in Chapter 4, that we should desire to have whatever is good to have, and that we should desire not
to have whatever is bad to have. That way our desires will be properly attuned to the world. There is something less
than perfect about us if there is something good to have, especially something about which we know, and we do not
want to have it. But if everyone

156

THE NECESSARY EVILS

desires what is good to have, since some good things can (of logical necessity) only be possessed by one person (as I
pointed out in Chapter 7) there will be the natural evil of many frustrated desires. Power over some individual area is
just such a good: if you are President, I cannot be President too. And monogamous long-term marriage to a particular
individual is another such good. Because being President of the United States would be a good thing for many people,
it is good that many people should desire it (though not bad if they lack the desire); and so inevitable that almost all of
them should suffer from frustrated desire. And because being married monogamously for life to John is good for
many women, it is good that many women should desire it; and so inevitable that almost all of them should suffer
from frustrated desire. If we were so made that we only desired what we could get, our desires would not be
responsive to what is good in things. So it is good in one way that we should be bored by what is uninteresting,
saddened by our failures, mourn the loss of loved ones. Then our emotional nature pays proper tribute to what is good
and bad in things.

One of the deepest of all our responsibilities for our fellows, as for ourselves, is the responsibility for life or death. God
could have made a world in which all creatures unavoidably live for ever. But in such a world we would not have the
very deep responsibility for how long others live. Our world is one in which we can do to each other the worst of
wrongs, for we can deprive each other of the greatest of goods — life itself. It is always a mark of supreme trust to allow
someone to have a gun. If God made this world, he made a world with many guns — many ways of depriving others of
life, and many ways of allowing others to die through negligence; and therefore also a world with many ways of acting
to keep others alive. This is a world with much scope for saving individuals in the short term (e.g. rescuing them from
drowning), saving individuals in the longer term (paying for their medical care), and acting so as to save many future
individuals (founding hospitals, and making non-polluting motor vehicles). Among the ways in which the choice of
killing or saving the life of others is available to us is by our having the choice of preventing or not preventing the
suicide of others. And the choice available to us here is a choice of how much responsibility we are willing to give to
others. It is no doubt good to stop others acting hastily under the influence of a desire which they desire or think it
good not to have; but bad to stop someone exercising

THE RANGE OF MORAL EVIL

157

an ultimate freedom whose consequences he desires, and desires to desire, and has thought through. But once again in
all these ways much responsibility given to free agents will make it enormously probable that there will be much moral
evil unprevented by God.

Human choices, I have emphasized, may reinforce each other and have long-term consequences; and it is good that
they should. The possibility that our bad choices will cause the suffering of victims distant in time and space (which
God is unlikely to prevent) gives us a yet greater responsibility in the choices we make. The suffering and deaths of the
Jewish victims of the Nazi concentration camps were the result of a web of bad choices stretching back over centuries
and continents. So many humans spread false rumours about Jews, developed anti-Semitic propaganda without
considering counter-arguments, limited the employment and educational possibilities for Jews, confined Jews to
ghettos, and so on, until Hitler was able to issue orders to exterminate Jews which had some prospect of being carried
out. And the sufferings and deaths in the concentration camps have in turn caused or made possible a whole web of
actions and reactions stretching forward over the centuries of sympathy for victims, helping their relatives (to set up
the state of Israel), avoiding any such event ever again, etc. The possibility of the Jewish suffering and deaths at the
time made possible serious heroic choices for people normally (in consequence often of their own bad choices and the
choices of others) too timid to make them (e.g. to harbour the prospective victims), and for people normally too hard-
hearted (again as a result of previous bad choices) to make them, e.g. for a concentration camp guard not to obey
orders. And they make possible reactions of courage (e.g. by the victims), of compassion, sympathy, penitence,
forgiveness, reform, avoidance of repetition, etc., by others. (On the goodness of different kinds of reaction to
suffering, see the next chapter.) Of course, I am not saying that anyone other than God would have the right to allow
such things to happen, without intervening to stop them. (On God's right, see Chapter 12.) And, as I am emphasizing
throughout, there is obviously a limit to the moral evil which God will allow us to cause (as there is to the natural evil
by which he will allow us to be afflicted), but it is not obvious where that limit lies. And note again that the suffering in
the Nazi concentration camps was the result of a very large

158

THE NECESSARY EVILS

number of free bad choices over many centuries, and made possible very many future good choices.

Incentive and Deterrent Desires

If ignorance about the consequences of our actions is a good thing, in order that we may have the choice of whether or
not to acquire knowledge, and ignorance by some is a good thing in order that others may have the choice of whether
or not to give them knowledge, there is clearly a risk of us harming ourselves without realizing that we were going to
do so. And even if to some extent we do realize the consequences of our actions and realize that they are bad, the bad
desires may sweep us on to act badly in such a way that it would take a very strong and very quick act of will to stop
them. We need time for serious consideration of alternatives.

In considering what a God will do when he sees creatures knowingly or unknowingly taking steps which lead to
disaster, we have an obvious analogy in the case of the human parent in relation to her children. In the case of some of
their actions children, especially very young children, are often quite unable to understand the consequences, even if
the parent tells them (the consequences, for example, of running into the road after a ball or not taking medicine). And
in the case of other actions, although a child may understand the consequences, his will may be weak: he may not clean
his teeth or do his homework, through sloth. What does the good parent do in these situations? The parent provides
incentives and deterrents. In the case of the very young who cannot understand the consequences of their actions, the
parent ties the child's action to consequences which he can understand. She may provide an immediate deterrent by
trying the action to a corporal punishment: Tf you run into the road, I will beat you.’ Or an immediate incentive by
tying the unwelcome medicine to a welcome food (‘sugaring the pill’). With respect to how to deal with the actions of
somewhat older children whose will is nevertheless very weak, the parent is inevitably torn. She is anxious for her
children to grow up to have increasing freedom to choose their own lives and work out what is best for themselves.
Too much interference, especially as a child grows older, would be bad. Yet, aware that the child's will is weak,

THE RANGE OF MORAL EVIL

159

the parent wishes to encourage him to do what is right, but not to force him. So here too she will to a limited extent
reinforce the child's will with incentives and deterrents, but typically delayed incentives and deterrents (see Chapter 4)
to a continuing course of conduct which give the child time to make a considered judgement about what to do;
rewards for doing the homework regularly, more distant punishment for not cleaning teeth regularly. A good parent
steers the middle course between leaving everything to the child on the one hand and on the other hand dictating what
the child is to do.

Analogy suggests that a good God will behave like a good parent. He will provide immediate incentives and deterrents
to do actions needed for life and growth into undamaged organisms. There are the incentive desires for food, drink,
sex, and sleep; and the deterrent desires to avoid the pain of the fire, or the danger of heights or dark. And as we come
to know how the world works, he provides the delayed incentives and deterrents to courses of conduct which give us a
chance to choose our course of action in a more considered way. When we come to know that frequent smoking
causes cancer or that considerate courting will gain sexual pleasure, we are provided with deterrents and incentives to
actions good for us and others for other reasons — our survival and that of our race.

Often, however, our only way of coming to know that some course of action will lead to a lengthy frustration of desire
(a pain we cannot immediately remove, an urge we cannot immediately satisfy) is by previous experience. I know that
drinking more than eight pints of beer will cause a hangover, because it has done so several times before. I shall be
arguing in Chapter 10 that, unless we learn that way, further good things will be lost.

The biological incentives and deterrents which lead us to eat and drink the right things and avoid the dangers of injury,
heights, and the dark are reinforced by the desire to be thought well of, especially by those who love and care for us;
and the desire to give pleasure and not sadness to those who love us, which leads to the very young accepting guidance
to do certain things and avoid other things, when they cannot understand why it is good to do so. The latter great
desire is operative to deter agents from doing so many harmful things — and that includes deterring them from suicide.
The desire to live is almost always a very strong one, and the would-be suicide who thinks that there is no good in his
staying alive has an initial strong counter-inclination to overcome. And

160

THE NECESSARY EVILS

some of the kinds of suicide which I mentioned earlier look pretty painful or drawn out, and that also deters. But the
desire not to hurt those who love us is also a powerful influence. It is the influence of love.

The biologically useful desires which form the nursery safety barrier are not of course always strong or relevant
enough to keep even babies safe. But if they were, then parents would have no responsibility. As children grow older,
they can choose by yielding to strong contrary desires to overrule the biologically useful desires; to force themselves, in
order to win a bet, to keep their hand in the fire, say. And, a major theme of this book is that only if agents have the
power to do bad do they have significant responsibility. True, I am suggesting both that it is a good thing that agents
have bad desires, and that it is a good thing that agents have good desires which provide a counter-balance, at any rate
initially, to indulging those bad desires. The former are necessary for a situation of temptation at all. The latter are
necessary to prevent things going awry too soon or too often, with immediately disastrous consequences. True, there
are not desires to help us to avoid all disasters and seek all goods, for, if there were, there would be no incentive to find
out the consequences of our actions. But the simple deterrent and incentive desires which we do have are a useful
protection which a caring God would provide if he was interested in our simple pleasures and continuing life as well as
in our having deep responsibility.

Our best desires may sometimes get crowded out; we ignore the long-term issues of the sort of person we choose to
be in favour of putting all our energy into fulfilling other desires, good but less good; desires for instant sensory
gratification, or for fame and fortune. While it is good that we shall be allowed to choose in the end whether to be the
sort of person whose desires for the bad or less good are stronger than the desire to form a good character, this is a
momentous choice which it is good that we should have time to make. It is therefore good that God should provide
some mechanism to help us in this situation. One mechanism which he does provide is simply failure of an
irrecoverable kind. Failure in an exam which cannot be taken again, or evidendy final failure in a personal relationship,
give us an opportunity to rethink our priorities. But we are also made so that failure of a recoverable kind (in an exam
which can be taken again; or a bankruptcy from which we can

THE RANGE OF MORAL EVIL

161

be released), with the resulting frustration, pulls us up, dampens the immediacy of the drive, so that again we have time
to think. We have time to choose whether to allow ourselves to be the sort of person who is dominated by those
desires or gradually to make ourselves a different sort of person instead (the sort of person in which desires for fame
and fortune are only strong when compatible with better desires). It is good that we shall have the choice; and that
means strong desires for bad states or lesser goods. But good that the choice shall be taken deliberately over time, and
hence good that too quick indulgence in desires for bad states or lesser goods be met by temporary failures. How
fortunate we sometimes are when some of our moderately worthy projects fail.

To draw attention to the deterrent value of various kinds of suffering is inevitably to raise the question why the
suffering still occurs when it is no longer able to deter. Among the advantages of the pain caused by fire is that it leads
the sufferer to escape from the fire. But the pain still occurs when the sufferer is too weak or paralysed to escape from
the fire. So we ask, would it not be better if only those able to escape suffered the pain? But if that were the case and
known to be the case (and it would violate the Principle of Honesty if God did not in the end let it be known), then
others would know that it mattered much less that they should help people to escape from fire and that they should
prevent fire. And so the opportunities for humans to choose whether to help others and guard against their future
sufferings would correspondingly diminish; just as fit's all insured’ lessens the thiefs reason for not stealing: stealing
won't cause real hardship. And in general if God normally helps those who cannot help themselves when others do not
help, others will not take the trouble to help the helpless next time, and they will be rational not to take that trouble.
For they will know that more powerful help is always available. The choices of providing or not providing significant
help to others, which I considered earlier in this chapter and suggested that it was very good for us to possess, would
not be available to us.

One useful kind of deterrent desire is the desire to avoid punishment which an authority has vowed (see pp. 95-6) to
impose on those who commit certain acts. By making a vow, we saw earlier, the vow-taker makes it a good thing that
he should do what he has vowed, when otherwise it might not be good, and might even be bad that he should do that
act. The badness of the suffering

162

THE NECESSARY EVILS

involved in the imposition of punishment on someone who ignores a threat to punish whoever does a certain crime
(when that threat is intended as a vow) is a necessary condition of the possibility of deterrence by vow.

Sloth and Decay

So for various reasons it is good that we have good desires and good that we have bad desires, desires which by their
influence provide a situation of serious temptation. The desires are desires to affect ourselves, each other, and the
world in various ways. Now we embodied humans, animals, and the inanimate world could have been such as to be
fine for ever if left alone. Our bodies might have been subject to no disease, ever self-renewing; and the same might
have been the case for the animals and plants on which we depend for life; and for the inanimate objects we build to
help us. Plants might need no tending, animals no fattening, and washing-machines and computers, once built, might
never break down. We and the animals would be reasonably content. We could make things even better or harm them
in various ways — but that would require active intervention and so effort. Even if the world contained sudden death, it
would contain none of the infirmities of old age. Death would come suddenly and unavoidably. Although the Book of
Genesis implies that the Garden of Eden was not quite like this, it suggests that it was not far off ; 132 and so we may call
this the Eden world.

Now our world is not an Eden world. Every human, animal, plant, and human-made structure is subject to decay; and
this decay can cause misery, if not prevented or alleviated or repaired. The ancient philosophers saw decay as the mark
of the sublunary world. It is because our world is a world of decay and not an Eden world that inactivity is often bad
and quite often wrong; and so sloth, the inclination to inactivity, is a desire which provides us

Contrast Gen. 2: 15, which implies that even the Garden of Eden required some cultivation, with Gen. 3: 17—19, which implies that the world outside Eden needed a great
deal more cultivation. I should make it clear that I am not making a claim that the story of the Garden of Eden should be read completely literally; I am simply using the
convenient word ‘Eden’.

THE RANGE OF MORAL EVIL

163

often with serious and quite often with very serious choices. We need to act (not merely to stop ourselves acting) in
order to keep ourselves and each other healthy, fed, sheltered, etc. Sloth is a source not merely of temptation to the
lesser good of not improving things, but of temptation to do what is bad or wrong (by failing to act). For we wrong our
fellows (and our own animals) if we let them slide into illness, homelessness, incapacity, and so on; and we do what is
bad (and, if there is a God to whom we owe our life, what is wrong) if we let ourselves and our plant and inanimate
environment degenerate. In an Eden world if we did nothing none of that would happen; activity would be required
only to improve things. And if things work well anyway, it would not matter very much if we did not improve things. I
should add, in order to avoid misunderstanding, that of course even in our world it is sometimes a good thing to be
lazy. A moderate amount of satisfaction of any of all self-directed desires, including sloth, is, other things being equal, a
good thing. And sloth is instrumentally useful in that its indulgence enables us to regain our strength. But indulgence
beyond a moderate amount is undoubtedly a temptation, not merely to a lesser good but often to the bad or wrong. It
is because houses crumble, disease gets hold of us, accidents cripple us, that it becomes bad if we do nothing about
things; and frequendy wrong.

In an Eden world temptations to the bad or wrong would have to be mainly in the form of temptations to interfere
with the good rather than to let things go on as they were. Instead of being tempted not to do anything to prevent the
spread of some disease and eventually eliminate the disease, by setting up an inoculation programme, starting a
research institute, etc., the temptation would be to discover a new disease, manufacture the bacteria, and help them to
spread the disease; or actively to do something, maybe good in itself, which, to our knowledge, would cause the spread
of disease. But the temptation actively to spread a disease could only come from a desire to hurt, to cause suffering.
And even if the temptation came in the second form (e.g. as a temptation to make others better off in some way at the
cost of hurting many others), it would still come in the form of a temptation to do something active requiring effort
which would have a known bad consequence. What goes for this example goes generally. In an Eden world temptation
to do bad or wrong would come in the form of temptations to hurt or to do something active which would
(knowingly) hurt others.

164

THE NECESSARY EVILS

In our world, the world of decay, bad doing and wrongdoing can so often be achieved by doing nothing; and the
temptation to do nothing is a temptation to do something good in itself (the satisfaction of a self-centred desire for
inactivity). And in fact the temptation to which we are all ultimately subject is sloth. We may lose our appetite and our
sex drive; but whoever we are, either we are satisfying a desire to do nothing; or, if we are actively doing something, in
the end we desire to stop. Sooner or later we all feel tired. We need not have been made like this at all. We could have
been made so that we never felt tired, but fell asleep suddenly when our bodies needed to recharge. Or we could have
been made so that our bodies never did need recharging; there might be certain actions we could do if we chose, and
others which we could never do — and we could not achieve any more by pushing ourselves harder. But in fact we are
not made like this. In the end we all start to feel tired, and we have to push ourselves to continue.

In consequence of all this, in our world of decay, we have open to us both the actions of negligently allowing
something bad to happen and the actions of actively causing something bad in itself or causally connected therewith (all
of which actions may be wrong ones). It is good that we should have the opportunity of very serious choice (in our
actions towards others) without having to desire to hurt, i.e. to hate our fellows (or animals or the inanimate
environment), which is a horrible thing, or even to do anything active (anything which required energy) which will hurt
them; only to love ourselves too much. Causing people to suffer is a far worse thing than letting people suffer. And yet
it is good that we should have the opportunity to do wrong. Being subject to sloth puts us in that situation where we
can do wrong without actively causing it. But it is also good that some of us sometimes have the temptation actively to
cause wrong, for by resisting that temptation we make more difference to the world for good, redirecting its inbuilt
tendency. Our world, being a world of decay, provides abundant opportunity for wrongdoing merely through sloth.
But it is also a world where we are subject to some bad desires actively to cause what is bad. But blessed be God for
putting us in a world of decay, where so many of our temptations are temptations to idleness.

THE RANGE OF MORAL EVIL

165

Conclusion

So, to conclude this chapter, like free will itself, responsibility can come in degrees. We could affect only ourselves, or
make great differences to many others, including future generations. It is intrinsically good (good for us) that we shall
have much responsibility, and make significant choices between many good and bad alternatives. To have a serious
choice of doing the bad we must have a desire for what is bad. The desire may be a desire for something normally
good, but under the circumstances of the choice bad. It is, however, good that the serious choices should include the
choice of doing nothing. It is intrinsically bad (bad for us) that the bad consequences come about through our choice.
But the more freedom and responsibility we have, of logical necessity the more and more significant are the bad
consequences which will result (unprevented by God) from our bad choices; and so the more probable it is that many
such will result. Every slight addition to our freedom and responsibility increases slightly the probability of sadness and
pain; every slight diminution of the probability of sadness and pain resulting from human actions diminishes slightly
our freedom and responsibility.

9 Natural Evil and the Scope for Response

And as [Jesus] passed by, he saw a man blind from his birth. And the disciples asked him, saying, Rabbi, who did
sin, this man, or his parents, that he should be born blind? Jesus answered, Neither did this man sin, nor his
parents: but that the works of God should be made manifest in him. We must work the works of him that sent
me, while it is day: the night cometh, when no man can work. When I am in the world, I am the light of the
world. When he had thus spoken, he spat on the ground, and made clay of the spitde, and anointed his eyes with
the clay, and said unto him, Go, wash in the pool of Siloam (which is by interpretation, Sent). He went away
therefore and washed, and came seeing.

(John 9: 1-7)

Human Response — in Feeling and Action

My main concern in Chapters 7 and 8 has been to show that the great good of very serious human free will and
responsibility cannot be provided without (the very great probability of) much moral evil. In the next three chapters I
shall be concerned with natural evil (i.e. the bad states not brought about or allowed to occur by humans), and I shall
seek to show how it makes possible many of the good states which I described in Part II. 133 1 shall be arguing in this
chapter that natural evil extends enormously the range of good desires which agents have and may cultivate; and the
actions available to them and, in the case of free agents such as humans, the range of actions which they can freely
choose.

‘Moral evil ( malum culpae ) God in no way wills . . . but the evil of natu
(St Thomas Aquinas, Summa Theologae , la. 19. 9).

oral defect (or the evil of punishment) he does will by willing some good to which such evil is attached’

NATURAL EVIL AND THE SCOPE FOR RESPONSE

167

A particular natural evil such as pain makes possible felt compassion — one's sorrow, concern, and desire to help the
sufferer. It is good that if pain exists, compassion exists, whether or not it can lead to action. It is good that one feel
compassion for the suffering of those with whom one is involved (for spouse, children, parents, friends), but also for
those with whom one is not involved, in distant lands and at distant times. But of course, the objector will say, even if
pain is better for the response of compassion, better still that there be no pain at all. Now obviously it would be crazy
for God to multiply pains in order to multiply compassion. But I suggest that a world with some pain and some
compassion is at least as good as a world with no pain. For it is good to have a deep concern for others; and the
concern can be a deep and serious one only if things are bad with the sufferer. One cannot worry about someone's
condition unless there is something bad or likely to be bad about it. If things always went well with someone, there
would be no scope for anyone's deep concern.

It is good that the range of our compassion should be wide — extending far in time and space. Good for those who
suffered long ago that they are not forgotten; good for us to react with love to distant creatures about whom we know
little. One of the great benefits of television has been to tell us about the tragedies and triumphs of peoples distant in
space, so that they too can be within the beam of our concern, even if we can do nothing about them. And it is not
beyond the powers of science to provide us with similarly detailed information about the triumphs and tragedies of
peoples long ago.

The sorrow of one in a distant land who really cares for the starving in Ethiopia or the blinded in India is compassion
for a fellow creature, even though the latter does not feel it; and the world is better for there being such concern. And
although it is good if the sufferer is aware of compassion shown to him, there is peculiar merit in an act of compassion
which cannot be returned, just as there is peculiar merit in acts of generosity which cannot even be recognized as such
by the recipient, let alone acknowledged.

I wrote in Chapter 4 also about the goodness of the persistent desire which persists though temporarily frustrated. It is
good for animals to long for the return of the lost mate, or offspring. It is a proper response to what is lost. And again,
though it would be crazy for God to multiply losses to secure the good of persistent desire, it is nevertheless good that
we should each be able sometimes to

168

THE NECESSARY EVILS

manifest serious longing, which is the response of emotions to important things.

The supreme tribute of our desires to loss is grief, and though it could never be enough of a good that we should
evince grief to justify the permanent loss of a loved one, it is nevertheless a small good which, together with other
good states made possible by that loss on which I have already commented and others to which we will come in due
course, might make the loss overall justifiable. Bad states often serve many good purposes, none of which by itself may
give God enough reason for allowing the bad state to occur but together they may do so.

What is known as the ‘higher-order-goods defence’ draws to our attention the good of performing certain sorts of
good action, namely those done in the face of bad states, and of having the opportunity freely to choose to do such
actions. The structure of this ‘higher-order-goods’ defence is as follows. A particular natural evil, such as physical pain,
gives to the sufferer a choice: whether to endure it with patience, or whether to bemoan his lot. His friend can choose
whether to show sympathy towards the sufferer, or whether to be callous. The pain makes possible these choices,
which would not otherwise exist. There is no guarantee that the actions directed towards the pain will be good ones,
but the pain gives agents the opportunity to perform good actions. The good or bad actions which we perform in the
face of natural evil themselves provide opportunities for further choice, of good or evil stances toward the former
actions. If I am patient with my suffering, you can choose whether to encourage or laugh at my patience; if I bemoan
my lot, you can teach me by word and example what a good thing patience is. If you show sympathy, I have then the
opportunity to show gratitude for the sympathy; or to be so self-involved that I ignore it. If you are callous, I can
choose whether to ignore this or to resent it for life. And so on . 134 There are different special values in doing the
different good actions. Courage is doing good

134 Aquinas sees the central point of the higher-order-gpods defence: ‘Many goods are present in things which would not occur unless there were evils’ ( Summa contra Gentiles , 3.
71. 6). He gives the examples of patience in the face of persecution and justice in the face of wrongdoing — which are clear examples, in the spirit of the present chapter, of
evils logically necessary for goods. But his example of ‘in the order of nature, there would not be the generation of one thing unless there were the corruption of another 5
raises the question what good is served by the order of nature being so structured, an answer to which I sought to provide at the end of Ch. 8 .

NATURAL EVIL AND THE SCOPE FOR RESPONSE

169

actions when it is harder than normal. An agent evinces his most substantial commitment to the good when he does
such actions when it is hardest, when he gets no encouragement from the success of other plans, and things are
happening to him which he does not desire. He makes such a commitment when he shows courage of a certain sort.
And it is good that others should be involved with people at their most naked making the hard choices, for the others
then show sympathy when it is hardest (because there is nothing about the appearance or success of the sufferer to
make him attractive) and when it is most needed, doing what they can for him in all ways, including helping him to
make the right choices. Help is most significant when it is most needed and it is most needed when its recipient is
suffering and deprived. We are also involved with a sufferer at his lowest when we merely feel compassion, but we are
more involved when we actively help to do something about the situation, and do not merely have the appropriate
feelings. And in cases of both kinds, the sufferer often has the good of feeling and seeing the concern of others for him
when he is unattractive and unsuccessful. ‘Sorrow shared is sorrow halved’, says the proverb, and it is only doing the
sums in so far as they affect the original sufferer who knows of the compassion. If we add the benefit (not altogether
an enjoyed benefit) to the sympathizer, the sums may well sometimes come out level.

But could not the absence of a good (of an ability to walk, say, or the ability to talk French) give to the victim equal
opportunity: whether to endure it with patience, or whether to bemoan his lot; and to friends, whether to show
sympathy or whether to be callous? To answer this question, it is important to recall (see p. 68) why pain is a bad state
and so, if uncaused by humans (and not negligently allowed to occur by them), a natural evil. Such pain is a natural evil
because it is a sensation strongly disliked. Any state of affairs believed to hold, not caused (or negligently allowed to
occur) by humans, disliked as strongly, would be just as bad. Some people dislike their disabilities just as much as they
dislike pain; they so dislike their inability to walk that they will undertake a programme to conquer it which involves
their ‘overcoming the pain barrier’. True, it would be unusual for anyone to dislike anything quite as much as some of
the pains caused by disease or accident (and to call those pains ‘intense’ just is to say how much they are disliked). And
for that reason, pain normally provides more opportunity for

170

THE NECESSARY EVILS

evincing patience rather than self-pity, despite contrary temptation, than does anything else. But any state of affairs
(believed to hold) disliked as much would be equally bad and so provide as much opportunity. And the choice between
being sympathetic rather than callous matters more then than it does if the suffering is less. If the absence of the good
is not disliked nearly as much as the sensations caused by disease and accident, then of course it is still very good to
show courage in seeking to remove that absence, but the courage is not in the face of such strong dislike for the
existing state of affairs.

The mutual involvement for which suffering provides opportunity is an involvement which more than one other can
have with the sufferer. Suffering provides the opportunity for chosen cooperation in mutual involvement. That is
especially so with respect to the good of cure and prevention. Pain often needs more than one doctor to cure it; it
needs a team of nurses and chemists. And the prevention of pain often needs the cooperative research of generations
of research workers, backed by the money provided by taxes for which political parties have fought in elections, or
which charity workers have gathered in collecting boxes over many years. And the frustration of desire which is mental
distress at deprivation of some kind provides further opportunity of a different and deeper kind for chosen
cooperative effort: cooperative political and charitable effort to help the exiled, the disadvantaged minorities, the
victims of child abuse, divorce, murder, or theft — systems of judges and lawyers, social workers and counsellors. And
with respect to cooperative effort, as with respect to individual effort, there is good in involvement with people at their
most naked, for the same reasons. Whatever it is good that we do, it is good that others help us to do, even if
sometimes it is also good that we have available the even harder choice of showing courage on our own. The bad state
of pain (or other suffering) is the grit which makes possible the growth of the pearl. Once again of course there clearly
comes a point where additional pain is a worse state than the good possibilities for courage and sympathy which it
allows. But it does not follow from that that some pain is not on balance a good thing for this reason.

Showing sympathy and courage are good actions which may be done in the face of the simple natural evils of pain and
other suffering. Other good actions such as making reparation and forgiving are done in the face of moral evils. But, in
these latter cases, however

NATURAL EVIL AND THE SCOPE FOR RESPONSE

171

the sums add up (whether the good of reparation and forgiveness ever outweighs the bad of the wrongdoing), it
cannot provide justification for the agent's wrongdoing; for wrongdoing would not be wrongdoing if the agent was
justified in so acting for the sake of a greater good . 135 And clearly sometimes there is wrongdoing. But that does leave
open the possibility that though the agent ought not to have done the wrong, a balance of good resulted from his doing
so. There are, too, good actions of certain kinds which can only be done in the face of good actions of various kinds,
such as showing gratitude, showing respect (e.g. in recognition of achievement), and reward; and the possibility of
these responses to actions which may themselves be responses to pain and suffering (e.g. showing gratitude to doctors
who have worked hard to relieve pain) provides further reason for permitting the pain and suffering.

There are, however, two kinds of reason why, it may be suggested, the choice of showing courage, sympathy, etc. might
be made available without the existence of natural evil. First, for most of these choices, apparent suffering, not real
suffering, is all that is necessary to provide the relevant opportunity. While an agent can only be courageous in the face
of his suffering if he really is suffering — for the reason that necessarily if we suffer we believe that we are suffering, and
conversely- — an agent can show compassion for the suffering which he believes another to be suffering, without that
other actually suffering anything. But for God to bring it about generally that humans seem to be suffering when they
are not really would be to practise a large-scale deception on the human race which would violate the Principle of
Honesty. It would be wrong of God to create a world of unfeeling robots surrounded by a few human beings who are
deceived into supposing that the robots suffer a lot, and so are moved to show sympathy and generous care when
there is not the slightest need for it. The only morally permissible way in which God could give the opportunity of
exercising such higher-level virtues as compassion is by actually allowing or making others suffer.

It may, however, be suggested, secondly, that adequate opportunity for the higher-level virtues would be provided by
the occurrence of moral evil without any need for suffering to be caused by natural processes. You can show courage
when threatened by a

See St Paul: ‘Shall

in sin that grace may abound? God forbid’ (Rom. 6: 1-2).

172

THE NECESSARY EVILS

gunman, as well as when threatened by cancer; and show sympathy to those likely to be killed by gunmen as well as to
those likely to die of cancer. I commented at the end of the last chapter on how a world of decay gave us the
opportunity actively to prevent disease and other malfunctioning, or to neglect to do so. That opportunity, which arises
only in so far as we know how to reverse the decay, is an opportunity to do or avoid moral evil. My concern here is
with the suffering which we do not know how to prevent and so constitutes natural evil (some but not all of which may
be caused by decay; for example, the effects of earthquakes are not a matter of decay), and for the opportunities for
response which it gives us. The objection is that all these opportunities could be provided by human malevolence
(actively causing suffering) or negligence (allowing it to occur). Mere malevolence would not provide much opportunity
for courageous response, but the possibility of negligence allows more.

Yet just imagine all the suffering of mind and body caused by the disease, earthquake, and death not immediately
preventable by humans, removed at a stroke from our society. No sickness, no unavoidable diminution of powers in
the aged, no birth deformities, no madness, no accidents. Then not merely would none of us have the opportunity to
respond with sympathy or courage or reforming zeal etc. directly, but so much of the oppression of one group by
another stimulated by such suffering would also be removed. Starvation and disease of one group have so often served
as the triggers leading to their oppression of another group whose good things they sought to possess. (That is, those
natural evils have so strengthened the desires of the former for food and easier living that they yielded to them despite
their dim awareness that they had no right to oppress the other group.) Then so many opportunities for coping with
difficult circumstances would have been removed that many of us would have such an easy life that we simply would
not have much opportunity to show courage or indeed manifest much in the way of goodness at all. It needs those
insidious processes of (currently) unavoidable accident and dissolution which money and strength cannot ward off for
long to give us the opportunities, so easy otherwise to avoid, to become heroes. True, God could compensate for the
absence of natural evil by subjecting humans to such temptation deliberately (or at any rate knowingly) to cause
suffering to each other that there was again

NATURAL EVIL AND THE SCOPE FOR RESPONSE

173

plenty of opportunity for courage. He could make us so naturally evil that we lacked much natural affection and had
inbuilt urges to torture each other (or at any rate allow each other to suffer), in face of which we others could show
courage and sympathy. But it is, I hope, in no way obvious that it would be better for God to replace disease by such
an increase of inbuilt depravity (i.e. a system of strong desires for what is known to be bad or to cause what is bad).
Rather, I would have thought, the reverse. A world in which humans (and animals) lacked much natural affection for
parents, children, neighbours, etc. would be a horrible place.

So by permitting (by bringing about) the natural evil of physical pain and other suffering God provides a bad state such
that allowing it, or an equally bad state, to occur makes possible and is the only morally permissible way in which he
can make possible many good states.

All the actions with which I have been concerned in this chapter, of showing courage, sympathy, generosity, etc., are
actions of benefiting others. But I now apply to these examples points which I made in Chapter 5. It is a great good for
the agent if he can help someone who needs help. He is privileged to have the opportunity to be of use and blessed if
he takes it. God does a great good for us if he gives us such opportunity. He can only do this by building a world in
which natural processes ensure that by our actions we can bring benefits to others which they cannot easily secure in
any other way. But there are also benefits for the one who needs help. He is also privileged to be the vehicle who
presents the one who can help with the opportunity to help; he makes the life of the helper matter, by giving him a
serious choice whether or not to help someone who needs help. And if help is provided, then of course the sufferer is
further blessed by someone caring for him. My pain gives you the opportunity to show sympathy and help to relieve it.
You are fortunate to have that opportunity. I am fortunate to be able to give it to you; and doubly fortunate if you
mind and get involved with my need.

Human Response in Forming Souls

It is good, as we have noted earlier, that we should have the opportunity over time freely to form our characters, to
determine the

174

THE NECESSARY EVILS

kind of people we are to be. One aspect of character is moral belief. As I argued earlier, beliefs are a passive matter: we
do not choose our beliefs; they are forced upon us by our experience of the world. But what we can choose to do is
seek out relevant experiences, think through their consequences for morality, and so acquire new moral beliefs (not
ones which we choose in advance but whatever ones new experience forms for us).

Given the goodness of natural sympathy, a natural desire for the well-being of others and sorrow at their suffering, it is
clearly good that moral growth should be advanced by experience of the suffering of others, sympathy with whom
leads to deeper understanding of wherein lies their good, rather than by mere abstract thought. And that, of course, is
what often happens. Those with experience of particular situations (rape, abortion, marriage breakdown, etc.) come to
change their views about the morality of actions which involve these, through sympathy with those concerned. And
when we have not ourselves had such experience we can freely choose to seek out those who have before coming to
form a view about the moral principles involved. The suffering becomes the tool which we use for our growth of
moral understanding, and so in yet another way the sufferer is of use to us in helping us so to grow. I think that moral
growth could be attained other than by experience of suffering, but because of the goodness of sympathy, it is good
(good for the others) that the suffering of others should be of use in this way. Also, in enabling one to detect the good,
sympathy acts as a countervailing desire to pride in the correctness of one's own views. It is, however, important to
note in this connection, sympathy with sufferers is not going to be enough to enable someone to form correct moral
views about a situation. Abstract and disinterested thought has also its role to play in moral growth. Those who have
had experience of and reflected on only a narrow range of situations may be blind to moral distinctions to which they
would become sensitive by a wider knowledge of the world, and by reflection on the connections between situations.
Someone who reflects only on the suffering involved for some wife in an unhappy marriage may fail to see the
suffering which would be involved for the children in a divorce and the influence which such a divorce would have on
others with marital difficulties; and failing to see these things would fail to see what would be wrong in a wife getting
divorced. Whether humans

NATURAL EVIL AND THE SCOPE FOR RESPONSE

175

choose to reflect on the connections between situations is a matter for their free choice.

But of course character is not just (or even largely) a matter of true moral belief; it is largely a matter of desire, of the
natural inclinations we have to respond to situations. And while it is good in itself that we should have some
opportunity to exercise the higher-level virtues of showing courage, sympathy, etc., it is also good that we should have
the opportunity significantly over time to form our souls so that we are the sort of persons who naturally show
courage, sympathy, etc. It is good that that choice itself requires hard work in order that our commitment be genuine.
As I emphasized before, it is a very important contingent truth about humans (and animals) that by doing actions of a
certain sort when it is difficult, we make ourselves the sort of persons who do such actions readily. So showing courage
etc., which can only be done in the face of suffering, has a dual role. It is good that we show courage, and it is good
that thereby we make ourselves naturally courageous. Humans only have a really good character if it is the sort of
character which responds readily to suffering (in others and in oneself) in the right way. Natural evil provides the
opportunity not merely to be heroic, but to make ourselves naturally heroic. Without a significant amount of natural
evil, we simply would not have the opportunity to show patience and sympathy on the heroic scale required for us to
form heroically good characters. It is a great good for us to be able, through free choice over time, to form such
characters.

The formation of our character through our own free choices which I have emphasized to be so good a thing can only
be made by choosing to do the (believed) good rather than the (believed) bad or wrong. But many, as a result of their
own bad choices or the bad environment (in the formation of which the bad choices of others will have played their
part), do not recognize much bad or wrong as bad or wrong. Only the starkest and most horribly wrong acts do they
recognize as such. They have not yet reached but are close to the brink of total insensitivity to moral goodness. Their
only hope is to be presented with stark choices — evils which even they can recognize; and a God as concerned for
their salvation — i.e. their becoming good people (through their free choices) — as for the salvation of near-saints will, if
compatibly with his goodness he can, provide them with such choices. That means giving them the opportunity to
resist temptations to do cruel acts. He will give them

176 THE NECESSARY EVILS

the opportunity to feed the dying beggar rather than leave him to starve, to disobey orders to kill a Jew imprisoned for
his race, or not to torture animals. Yet of course those choices are only available if these terrible evils will happen if the
chooser refuses the right choice. And there must be, as there is, a limit to the suffering which God allows anyone to
endure for the sake of such a great good as the salvation of the hard-hearted. But it remains the case that it is a great
privilege for anyone to be the means of making available serious choices, even or possibly especially for the hard-
hearted.

That suffering is a blessing for the sufferer in the opportunities which it provides him for heroic action and character
formation is a constant theme in the spirituality of much Eastern Christianity, and especially of that compilation of the
spiritual writings of the Eastern Fathers the Philokalia, which has, more than any other book, formed the spirituality of
the Eastern Orthodoxy of the last two centuries. Take, for example, the passages which it includes from St Peter of
Damascus:

Through what are regarded as hardships we attain a state of patience, humility and hope of blessings in the age to
be; and by these so-called hardships I mean such things as illness, discomfort, tribulation, weakness, unsought
distress, darkness, ignorance, poverty, general misfortune, the fear of loss, dishonour, affliction, indigence, and so
on. Indeed, not only in the age to be, but even in this present age these things are a source of great blessing to us . 136

These ‘gifts’, as St Peter calls them, allow us (but do not compel us) to respond in the right way to them, by actions of a
kind which we would not otherwise have the opportunity to do. The point of poverty is that ‘one can endure it with
patience and gratitude’; the point of sickness is ‘so that one may earn the crown of patience’; the point of unsought loss
of goods and possessions is ‘so that one may deliberately seek to be saved and may be helped when incapable of
shedding all one's possessions or even of giving alms’. And these gifts provide opportunities for others to respond in
the right way, and a point of the opposite gifts is to enable them to do so. Wealth, writes St Peter, enables us to
‘perform acts of charity’; and he comments later that ‘without the poor’ the rich cannot save their souls

136 G. E. H. Palmer, P. Sherrard, and K. Ware (ed. and vans.), Philokalia, til (Faber & Faber, 1984), 172-7 . For an example from a work w
see St John of Damascus, On the Orthodox Faith , 2. 29 on the diverse goods which diverse bad states make possible.

vhich is not included in the Philokalia ,

NATURAL EVIL AND THE SCOPE FOR RESPONSE

177

or flee the temptations of wealth. Health enables us to ‘assist those in need and undertake work worthy of God’. And
so on.

To repeat myself so as to avoid any possible misunderstanding, wealth and health must be good for other reasons also,
e.g. for the simple sensory pleasures they can give their possessors; otherwise there would be no point in seeking to
pass them on to the poor and diseased. And of course poverty and disease are bad states, but they do serve the good
purposes which St Peter states, among other good purposes; and that provides reason for a good God to give them to
us for a limited period.

Animal Response

It is good that the intentional actions of serious response to natural evil which I have been describing should be
available to simple creatures lacking free will. As we saw earlier, good actions may be good without being freely chosen.
It is good that there be animals who show courage in the face of pain, to secure food and to find and rescue their
mates and their young, and sympathetic concern for other animals. An animal life is of so much greater value for the
heroism it shows. And if the animal does not freely choose the good action, it will only do the action because on
balance it desires to do so; and when its desire to act is uncomplicated by conflicting desires, the good action will be
spontaneous. And (even if complicated by conflicting desires), animal actions of sympathy, affection, courage, and
patience are great goods.

Yet an animal cannot go on looking for a mate despite failure to find it unless the mate is lost and the animal longs for
it; nor decoy predators or explore despite risk of loss of life unless there are predators, and unless there is a risk of loss
of life. There will not be predators unless sometimes animals get caught. A hunt would be only a game unless it was
likely to end in an animal getting caught and killed; and animals would not then be involved in a serious endeavour.
And there will not be a risk of loss of life unless sometimes life is lost. Nor can an animal intentionally avoid the danger
of a forest fire or guide its offspring away from one unless the danger exists objectively. And that cannot be unless
some animals get caught in forest fires.

178

THE NECESSARY EVILS

For you cannot intentionally avoid forest fires, or take trouble to rescue your offspring from forest fires, unless there
exists a serious danger of getting caught in fires. The intentional action of rescuing, despite danger, simply cannot be
done unless the danger exists and is believed to exist. The danger will not exist unless there is a significant natural
probability of being caught in the fire; and to the extent that the world is deterministic, that involves creatures actually
being caught in the fire; 137 and to the extent that the world is indeterministic, that involves an inclination of nature to
produce that effect unprevented by God. (And, I shall argue in Chapter 10, the danger cannot be believed to exist by
animals or — barring a crucial qualification — humans, unless the animal or human has observed creatures actually
being caught in fires. Fawns are bound to get caught in forest fires sometimes if other fawns are to have the
opportunity of intentionally avoiding fires, and if deer are to have the opportunities of rescuing other fawns from fires.)

True, the deterministic forces which lead to animals performing good actions sometimes lead to animals doing bad
intentional actions — they may reject their offspring or wound their kin — and in this case the bad action cannot be
attributed to free will. Nevertheless, such bad actions, like physical pain, provide opportunities for good actions to be
done in response to them; e.g. the persistence, despite rejection, of the offspring in seeking the mother's love or the
love of another animal; the courage of the wounded animal in seeking food, especially for its young, despite the wound.
And so on.

It is a mistake, in my view, to regard the killing of one animal by another for food as in itself a natural evil. To be killed
and eaten by another animal is as natural an end to life as would be death by other natural causes at the same age. For,
given that animals do not have free will or moral concepts, the actions which they perform do not have a moral
character, and so the killing of one by another is as much part of the natural order as is accident or disease. And if
death by such natural causes is not as such an evil, as claimed in Chapter 6, but simply the end of a good, so too with
death by predator.

If the behaviour of tossed coins is deterministic, talk about a natural probability of a coin landing heads can only be intelligibly construed as talk about proportions of coins
tossed in typical set-ups which result in heads. (‘Natural probability* or ‘physical probability* is probability in nature in contrast to ‘epistemic probability’, which is probability
relative to our knowledge.) See my An Introduction to Confirmation Theory (Methuen, 1973), chs. 1 and 2 .

NATURAL EVIL AND THE SCOPE FOR RESPONSE

179

Natural evil comes in only in so far as there is pain involved in the killing, or offspring who need parents are knowingly
deprived of them.

So given all this, is the opponent of theodicy really right to insist that the world would be better without the challenges
to and courage shown by animals? I do not think so. The world would be much the poorer without the courage of a
wounded lion continuing to struggle despite its wound, the courage of the deer in escaping from the lion, the courage
of the deer in decoying the lion to chase her instead of her offspring, the mourning of the bird for the lost mate. God
could have made a world in which animals got nothing but thrills out of life; but their life is richer for the complexity
and difficulty of the tasks they face and the hardships to which they react appropriately. The redness of nature fin tooth
and claw’ is the red badge of courage. In the absence of humans intentionally causing animal suffering, animals could
not do these significant actions without there being natural evils to which to react. And it would not be better if the
possibility of animals doing these actions was brought about by humans causing animal suffering, for humans have no
right to cause animals to suffer. God alone, as we shall see in due course, has that right. And if he exercises it, he makes
it possible for animals to do things that matter. But, as ever, there is the price to be paid for this possibility; and if
nature would not obviously be better without it, it would not obviously be worse. There is something to be said for the
lion lying down with the lamb.

The Extent of Animal Suffering

It is important to say something at this stage about the extent of animal suffering. Just how far down the evolutionary
scale sentience, and so suffering, goes is something about which we must inevitably be uncertain. For the grounds for
ascribing sentience to animals are the similarity of their behavioural responses to the stimuli which induce pain in
ourselves, and the similarity of the organization of their brains to our brains, which are the vehicles of pain in ourselves.
As we move down the evolutionary scale we come to animals less and less like ourselves in these respects; but there is
no obvious place at which we should conclude that sentience

180

THE NECESSARY EVILS

begins. Since behaviour and brain states are but fallible evidence of mental states, our uncertainty on this matter is
unavoidable. That said, I suggest that there are only weak grounds for attributing sentience to invertebrates (to ants
and crabs and moths), for they do not have anything like our sort of brain. When conscious life began, it probably
began with slight sensations; and then as it became more complicated, purposes, desires, and beliefs of the simplest
kinds, and finally the more sophisticated kinds of these (such as the cat's desire for knowledge), would have emerged.
The first feelings of pain in animals lacking belief must have been very slight in comparison with our pains . 138 For if at
some stage there was no feeling and now with humans there is much, and all the indicators of mental life show a
progressive development in intelligence, belief patterns, and other aspects of the mental life, we would expect a similar
growth in the intensity and complexity of suffering. The lower mammals, I therefore suggest, suffer very much less
than humans do. Further, for creatures without any beliefs, pain would be without significance; they would not be sad
because they had lost offspring, nor would they believe their physical pain to be inflicted by enemies. Nor would they
feel a pain as part of a lengthy period of pain; that would require memory of what had happened and knowledge of
what was going to happen. On the other hand, for such creatures who feel but do not believe and so do not perform
intentional actions, their pain could not be justified by any good for the suffering creature itself except the good of
being of use to others, e.g. by making available opportunities for proper response to it by others, opportunities to show
sympathy and to heal . 139 However long ago was the suffering of some child or animal, and even though no one at the
time knew about it, we humans can find out later and be sorry. Our compassion for sentient creatures is often far too
narrow; it needs to extend far over space and time.

138 Marian Dawkins suggests that since pain would only have given organisms liable to it an evolutionary advantage if it was correlated with risk of damage to the organism and
if the organism was clever enough to see how to escape from the suffering, i.e. if pain arrived on the evolutionary scene at the same time as beliefs and intentionality, it would
follow that without the latter, pain would not survive for long. She argues that in consequence very good evidence of when animals suffer is provided by whether they seek
energetically to avoid some state of affairs. See her ‘Minding and Mattering’, in C. Blakemore and S. Greenfield (eds.), Mindvaves (Blackwell, 1987) .

139 ‘One of the functions of pain, in species that are capable of it, is to awaken compassion’ (Austin Farrer, "Love Almighty and Ills Unlimited (Collins, 1962), 102) .

NATURAL EVIL AND THE SCOPE FOR RESPONSE

181

(I come to discuss the issue whether, good though it may be for the sufferers, God would be justified in allowing some
to suffer to give others such opportunities, in Chapter 12.)

Where animals do begin to have beliefs, then physical pain will be felt more and come to have a bit of the meaning
which makes it so much worse for us. But that is necessary if animals are to be able to perform those intentional
actions which this chapter has urged to be such a good thing. And of course as the capacity for more intense and
sophisticated suffering increases, so too does the capacity for more intense and sophisticated pleasure and other good
states. The higher mammals can enjoy play and courting; and they can long for things absent which they eventually
find. They can be inquisitive, and rejoice in the success of their activities.

10 Natural Evil and the Possibility of Knowledge

With most of [our fathers] God was not well pleased: for they were overthrown in the wilderness. Now these
things were our examples, to the intent we should not lust after evil things, as they also lusted. Neither be ye
idolaters, as were some of them; as it is written. The people sat down to eat and drink, and rose up to play.
Neither let us commit fornication, as some of them committed, and fell in one day three and twenty thousand.
Neither let us tempt the Lord, as some of them tempted, and perished by the serpents. Neither murmur ye, as
some of them murmured, and perished by the destroyer. Now these things happened unto them by way of
example; and they were written for our admonition.

(1 Corinthians 10: 5-11)

The Need to Know About the Effects of Our Actions, Good and Bad

For humans to have a choice between doing good and doing bad, we need to have true beliefs about the effects of our
actions, for the goodness or badness of an action is so often a matter of it having good or bad effects. It is bad to kick
other people because it will hurt them, good to give the starving food because that will enable them to stay alive. And
so on. So if God is to give us the choice between good and bad, he must give us, or allow us to acquire, true beliefs
about the effects of our actions — beliefs in which we have enough confidence to make it matter how we choose. We
need a whole sheaf of strong true beliefs with respect to many different actions, about what effects will follow from
them. How is God to

NATURAL EVIL AND KNOWLEDGE

183

give us these beliefs? Clearly we will have to think of our beliefs as justifiec^ (see p. 59), for if we do not think them
likely to be true, we will not really believe them. And so if God is to provide us with beliefs, he must, by the Principle
of Honesty, provide us in general with justifiec^ beliefs, for if our criteria of evidence were generally erroneous, we
would have no criteria by the application of which we could discover this, and so we would be systematically deceived
in innumerable ways about the world, and could not correct this.

Our justified^ true beliefs could only rarely depend for their justification on false beliefs unless the world is organized
on a systematically deceptive basis. If our true beliefs depended generally for their justification on false beliefs it is most
unlikely that they would often be other than false, barring the institution of a system designed to prevent this. For
example, it is most unlikely that erroneous reports of observations of the past would allow us to infer to a conclusion
about the future which turned out to be true, unless reporters were programmed to give exactly those erroneous
reports which allowed justified inference to true beliefs about the future. But if God created a world in which that sort
of thing normally happened, that massive deception would violate the Principle of Honesty. Suppose, for example, that
God created a world in which so far no one has taken heroin, yet, by making it the case that many observers
erroneously report cases of taking heroin leading to death, allows us to reach a strong justified true belief that taking
heroin leads to death. These observers would need to be systematically deceived if this sort of thing was to happen on
a regular basis. Further, we will only have justified beliefs as a general rule (as opposed to being the result of an
accident in particular cases) if there is a mechanism in place for ensuring that; that is, beliefs must be correctly ‘based’.
So if God is so to give us the strong true beliefs we need to make our serious choices, he must give us in general
knowledge of the effects of our actions: knowledge, that is, in the internalist sense, with which alone I shall be concerned
henceforward. 140

I first advocated the defence from the need for knowledge in ‘Natural Evil’, American Philosophical Quarterly, 15 (1979), 295—301 ; and developed it more fully in The Existence
of God (rev. edn., Clarendon Press, 1991), chs. 9, 10, and 11 (see esp. ch. 11) ; and in ‘Knowledge from Experience and the Problem of Evil’, in W. J. Abraham and S. W
Holtzer (eds.). The Rationality of Religious Belif (Clarendon Press, 1987) . It has come in for quite a bit of criticism. See David O'Connor, ‘Swinburne on Natural Evil’, Religious
Studies, 19 (1983), 65—73 ; Eleonore Stump, ‘Knowledge, Freedom, and the Problem of Evil’, International Journal for the Philosophy of Religion, 14 (1983), 49—58 ; Paul K.
Moser, ‘Natural Evil and the Free Will Defence’, International Journal for the Philosophy of Religion, 15 (1984), 49—56 . In my elaboration of it here I seek to defend it against these
criticisms. Those familiar with this discussion will recognize that I have amended the defence considerably in the light of these criticisms. It is no longer in the form of the
need for natural evil if we are to have knowledge of the consequences of our actions; but in the form of the need for natural evil if we are to have very well-justified
knowledge, and the opportunity to learn from experience and to choose to seek new knowledge.

184

THE NECESSARY EVILS

I distinguished in Chapter 1 between the necessary truths of morality, and the contingent truths which follow
therefrom when factual information about the effects and circumstances of actions is added. It is (plausibly) a
necessary moral truth that it is wrong to give money to beggars if they will spend it only on drugs which kill. It is a
matter of fact that certain beggars will spend money only on drugs which kill. It follows that it is a contingent moral
truth that it is wrong to give money to those particular beggars. Now I do not know of any good reason to suppose
that experience is necessary either for the possession or for the acquisition of concepts or knowledge of necessary
truths or their interconnection. Someone has a concept to the extent to which he can conceive what it would be like for
it to have application, and to the extent to which he can recognize that it does apply. And someone can see what is
involved in its application (i.e. know necessary truths which concern it) without ever having observed its application.
Someone might be born with an ability to conceive what it would be like for something to be red or green and to
recognize red and green objects, even if he has never observed such; and this ability would enable him to recognize as a
necessary truth that nothing can be red and green all over. The same applies to the necessary truths of morality.
Someone might know what wrong was when the world as yet contained none, and he might know which actions were
wrong and which states of affairs were bad before ever they had occurred. Our moral knowledge is not acquired in this
way but there is no reason why that of some human agent should not be. God could ensure that humans were given
moral concepts and a deep imagination which would enable them to comprehend necessary truths about their
application without their having any experience of harsh moral realities. We could know that it is good to feed anyone
starving even if we knew of no one who was starving; that torturing in order to extort belief in a creed is wrong, even if
no one had ever done it. However,

NATURAL EVIL AND KNOWLEDGE

185

I argued in Chapter 9 that although humans could acquire moral knowledge without experience, there was,
nevertheless, a certain value in their acquiring it through experience.

My concern here is not with knowledge of necessary moral truths, but with the factual knowledge of the effects of our
actions (the effects being described in non-morally loaded terms, e.g. just as pain or death) and so with the nature of
our actions (e.g. as causing pain or killing) which we need in order to know the contingent moral truths, e.g. that
certain particular actions are wrong (because they do have those effects). How is God to make such knowledge
available to us? The argument of this chapter is that while God might be able to give moderately well-justified
knowledge of the effects of our actions, good and bad, without too great a cost, he could not allow us to learn what the
effects are, let alone to choose to seek such knowledge, without providing natural processes (in which humans are not
involved) whereby those effects (good and bad) are produced in a regular way — or rather he could not do this without
depriving us of a very considerable other benefit. Natural evil is needed to give us the choice of whether to acquire
knowledge of the good and bad effects of our actions, and indeed in order to allow us to have very well-justified
knowledge at all.

God could perhaps implant in us strong true beliefs about the effects of our actions; or make us such that we gradually
find ourselves with more and more such beliefs as time goes by, beliefs which open up more and more possible actions
for us. We could start life with beliefs that crying causes adults to feed one, and kicking the bedclothes off causes one
to be cooler (and perhaps we do start life with those beliefs; however, I write ‘perhaps’ because it may be that when
babies want food or want to be cooler, they just cry or kick without having a belief that these actions will have the
desired effects). And then as we get older we could find ourselves with more and more complicated beliefs about the
effects of actions. Wondering how to hurt someone, I could find myself believing strongly that kicking him, or telling
others about the misdemeanours of his youth, would in different ways hurt him. I could perhaps find myself believing
strongly, for no reason at all, that setting light to hydrogen will cause an explosion, and that giving money to Oxfam
will relieve starvation, whereas spending it on buying books will not have this effect.

186

THE NECESSARY EVILS

If we are to have the opportunity of doing an action of any complexity over time, we will need to have beliefs at each
time as to which sub-actions in the particular circumstances of the time will contribute to the total action. In order to
have the choice of sailing round the world, I will need to find myself at each stage with a true belief about which
actions (tacking, going about, taking this course rather than that) will produce the sought-after result. And I will need
to find myself with beliefs about the effects of my actions on others of the crew. All that could happen. I could find
myself with the relevant beliefs. They might even be probabilistic beliefs — that if I pull the sail in further there is a 30
per cent chance that the boat will capsize. These beliefs, combined with the moral beliefs about which effects are good
and which effects are bad, would then allow us to choose between doing good and doing bad. Beliefs need never come
through experience of what we or others have done in the past or observed to happen in the natural world, or others
have told us; let alone as the result of our constructing a complicated scientific theory on the basis of many
observations.

Would these spontaneously arising beliefs amount to knowledge? I think that they might — just. I suggested in Chapter
4 that perhaps it is right for us to start from any beliefs (about contingent matters) with which we find ourselves; and
that would include any hunches about the effects of our actions such as those mentioned above. Such beliefs would
perhaps be justified; and the other conditions for knowledge (see earlier) would be satisfied. But these beliefs would
not be nearly as strongly justified and so amount as obviously to knowledge, as beliefs based on observation of what
has happened in the past and extrapolation therefrom. My belief B that if I set light to hydrogen there will be an
explosion is far better justified if it is not a basic belief, but justified by many beliefs to the effect that I have observed
that when I have set light to hydrogen in the past (or it has accidentally caught light) there has been an explosion. That
would seem to be involved in the necessary a priori standards of inductive inference (see Chapter 4) which upon
reflection we realize to be correct. For instead of a particular belief connected to the state of the world at the time it
concerns (the future) by an indirect causal chain, we would have many beliefs about past states of the world connected
to them by direct causal chains, from which our a priori standards allow us to make a strongly justified inference to the
future.

NATURAL EVIL AND KNOWLEDGE

187

An intermediate position is possible. B might not be justified by my beliefs about the past explosions which I have
observed, but simply by (true and justified) beliefs to the effect that I have always found in the past that my basic
beliefs about effects of my actions have subsequently proved true. But if I have always done actions with effects of a
certain kind — e.g. actions whose short-term effects concern only myself, or the physical well-being of those close to
me, or actions such that I have basic beliefs only about their good effects — there will remain a doubt about whether
my basic beliefs about the effects of actions with effects of other kinds are equally trustworthy.

If a belief that my action will have a certain effect is to be as well justified as possible, it will need to be backed up by
beliefs that in the past actions of just that type have had that kind of effect; or that the immediate result which the
action consists in bringing about has had that kind of effect in the past. What I mean by the latter is this. Every human
action done by means of the body consists in bringing about some bodily movement or immediate effect in the
environment, which in time has more distant effects. If I move my hand (intentional action), this consists of my
bringing about the motion of my hand; if I open the door, this consists of my bringing about the door being open. The
latter events — the hand moving or the door being open — I shall call results of my actions of moving my hand or
opening the door; they are events which could be produced by non-intentional causes. My belief that if I light hydrogen
it will explode will for its strongest justification depend on beliefs that in the past when I (or others) have lit hydrogen it
has exploded; or our beliefs that in the past when hydrogen has caught light accidentally it has exploded. By the
Principle of Honesty, such justifying beliefs must in general be true.

So if my beliefs about some contemplated action having a bad effect are to be as well justified 2 as possible, this will in
general require there having been similar actions, or events not produced intentionally, producing similar bad effects in
the past, which I can observe (or others can tell me about) and I can take account of. Given that it is good that our
actions shall be guided by our beliefs, it is good that we should have beliefs as well justified as possible. For in so far as
there is a justified doubt about which effects our actions will have, we can evade the moral force of a choice between
good and bad on the ground that it is not certain which effects our

THE NECESSARY EVILS

actions will have. The less certain the effects, the less serious the choice. If there is a doubt about whether smoking
causes cancer, it is less evidently a bad thing to smoke. To be well justified our beliefs need to be backed up by
experience. It may not be a necessary truth that all knowledge (of matter of contingent fact) comes from experience of
similar matters, but it is a necessary truth that it is better justified and so will give a firmer basis for action if it does.

But yes, perhaps, with this rather important qualification, we could simply find ourselves with knowledge of the effects
of our actions, well enough justified to provide a basis for considerable moral choice of which effects to cause. As
described it would not be a knowledge based on any understanding of how the world worked: I could just find myself
with the knowledge that to make this arm movement after this leg movement would enable me to sail round the world;
but I would not know how wind and tide and sail interact to make this possible. I could find myself with knowledge
about such matters also. Yet such knowledge would not be knowledge which I acquired by learning, let alone
knowledge that I chose to have and was prepared to sacrifice much to get. Knowledge would be ours unchosen.

Although I have done my best to describe this situation as one where humans would have something amounting to
knowledge (even if not knowledge as well justified as possible), it is difficult for me to avoid the feeling that my best is
not good enough. Could mental attitudes of mine really be described even as beliefs, let alone knowledge, about the
geography and history and science of the world if they did not result from experience, argument, challenge,
falsification? The mental attitudes which guided my behaviour in the absence of such empirical backing seem more like
instincts or hunches. Is that quality of reason which we value so much in humans really present at all in this situation?
If it is not, the need for natural evil will be even stronger than I am representing.

What, however, we could not do in this supposed situation of innate knowledge is to learn from experience, to discern
by observing its unintended consequences that some action of ours caused good or bad of some new kind and so to
learn that by doing a similar action we could cause good or bad. Above all, we could not choose to seek new
knowledge by thinking, searching, and asking — knowledge of what we can do and can bring about by our actions, as
well as theoretical knowledge of the structure of the world.

NATURAL EVIL AND KNOWLEDGE

189

We saw in Chapter 4 that the greatness of human (and to a lesser extent, animal) reason consists not only in having
knowledge, but in acquiring it and changing one's beliefs in the light of evidence. And we saw in Chapter 5 that it is a
great good for humans that we should have the free choice of whether to exercise our reason in finding out deep
theoretical truths about the universe — how big it is, how old it is, what it is made of, whether God sustains it, and so
on — as well as important factual truths about the human race and those close to us. And it is good also that we should
have the choice of whether or not to extend our power over others and the world; and extending that power involves
acquiring knowledge of which present actions will lead to humans being killed or fed, being educated or depressed,
going to Mars or living beneath the ocean. Not to have the choice of extending our knowledge of such matters is not
to have the choice of helping or harming people in new ways; and to refuse to exercise the former choice is not merely
to refuse to acquire the power to help or harm, but to risk helping or harming through the consequences of our
actions of which we are still unaware. Free choice without knowledge of effects is empty, and a significant kind of free
choice is the choice of seeking knowledge of how to bring about different effects.

Learning Involves the Operation of Natural Processes Producing
Good and Bad

Creatures who are to learn how to produce some distant effect must have a limited repertoire of basic actions that they
can do at will; and learning will consist of discovering that some basic actions done in certain circumstances will have
certain effects. Thereby we learn to produce those effects. We saw in Chapter 5 how having such a limited repertoire is
part of embodiment: I can move my hands at will, but produce more distant effects only by moving my limbs in
circumstances where those motions will have more distant effects. If God is to allow me to choose to acquire
knowledge of how to travel to Mars or to kill millions of people, he must allow me to learn which immediate states of
affairs (which could be results of my bodily movements) will lead to those effects. He must allow me to learn in which
circumstances will the depression of a button lead

190

THE NECESSARY EVILS

to a rocket going to Mars, and in which circumstances will it lead to nuclear explosions which will kill millions of
people.

Much acquisition of knowledge comes from being told by others, that is from testimony. But a belief of mine acquired
by testimony will only amount to knowledge if my informant or someone from whom he acquired his belief via a chain
of informants had a belief amounting to knowledge, whose strong justification was of a different kind. And if the chain
of testimony is too long, the justification which it produces for the resulting beliefs will be very much less. The human
race as a whole must learn in a different way.

All knowledge of the future is knowledge either of what natural processes will bring about or of what agents will bring
about intentionally (or both, if intentions are moulded by natural processes, or if they mould those processes).
Someone may infer to a future event either by regarding what will happen as to be produced by a natural process or as
to be produced intentionally. So knowledge that my action A, which consists in bringing about some result C, will have
a further consequence E, will be knowledge either that natural processes dictate that C brings about E, or knowledge
that some other agent on observing C will bring about E intentionally. Knowledge that putting cyanide in a man's drink
will kill him is knowledge of a natural process — that cyanide kills; and knowledge that you will visit me today when you
have promised to visit me is knowledge of intentional agency.

If God is to give to humans a range of actions with consequences bad and good, he must ensure that human actions
have these consequences, either as a result of a natural process which he implants in the world, or as a result of his
direct intentional action. And if the human agent is to learn about those consequences, he must learn about them either
by discerning the natural process or by discovering God's intention. God could allow me to learn the consequences of
my action by allowing me to ask him what he will bring about if I do the action. I might ask him, ‘How can I kill John?’
And then I might hear in my ear or see on the screen the English words ‘If you shoot John, he will die’. But if I regard
my actions as having the consequences they do in virtue of some other agent intentionally making the actions have
those consequences, I must regard that agent as in control of my life; and not merely my life, but, since he determines
the effect of my actions on others, as in control of their lives too. I must regard him as in control of the Universe, at
least

NATURAL EVIL AND KNOWLEDGE

191

locally. And I must regard him as perfectly good. For his local freedom of operation to determine what happens is (to
all appearances) absolute, and so therefore is his local knowledge of what will happen. And so, as the simplest
hypothesis, I must regard him as knowledgeable also in other fields, including morality, and free in other fields; and so
as knowing the good, and, not being distracted by temptation from pursuing it, as perfectly good. Under those
circumstances I could indeed discover the consequences of my actions, and know whether they are ones which I
believe good or ones which I believe bad. But I would regard my every movement as overseen by an all-knowing and
perfectly good being, i.e. a God. And this would be no mere balance-of-probability belief. It would be an evident belief
which guided every action of mine. That — I shall be arguing in the next chapter; and I must ask the reader to take this
result on trust until then — would make the choice between good and bad impossible, given that we have certain good
desires.

So, to preserve our serious choice between good and bad, God must implant in nature a system of natural causal
processes and let us learn what they are. I understand by a natural process one in which a cause of a given kind
produces an effect of another kind in a regular way either with natural necessity or with natural probability. Natural
processes are predictable processes; and if they are to be of any use to humans for prediction, the regularities must be
of a simple recognizable kind. The a priori principles of what is evidence for what outlined in Chapter 4 have the
consequence that the strongest evidence for the claim that Cs bring about Es will come from past observations of Cs
being followed by Es in varied conditions. If in the past mustard seeds being put in the ground and watered has always
been followed by the appearance of mustard seedlings, then very probably the implanting and watering of mustard
seeds causes the appearance of mustard seedlings. So the basic way in which God can allow us to acquire knowledge of
natural processes, that Cs cause Es, is by providing us many instances of the successions involved under different
conditions. He will implant in the world many instances of Cs being followed by Es under different conditions; and
allow us to reach our knowledge by reflecting on these observations, or — making it a choice which involves more
effort by us — allow us to search for and eventually find such evidence. Once we have this knowledge of causal
succession, then we

192

THE NECESSARY EVILS

know that if we produce a C as an immediate effect of a bodily movement, thereby we cause an E. Observing many
mustard seed-seedling successions, we come to know that the way to produce mustard seedlings is to sow mustard
seeds. We can make this kind of inference without automatically needing to suppose that God causes the system of
natural processes (although, I believe and have argued elsewhere, the operation of natural processes does in fact
provide an important part of a cumulative case for belief in God). And this of course is the way in which we do come
to learn about the effects of our actions, and see any beliefs about this we already have as strongly justified.

We (i.e. humans in general) learn that eating toadstools causes stomach pain by seeing people eat toadstools and then
suffer pain. We learn that alcohol makes people unsteady drivers by seeing people have many drinks and then drive
unsteadily, and so on. And we choose whether to acquire such knowledge, by choosing to search for and find
observational evidence from which we infer such causal processes. These observations open up a range of possible
actions, good and bad, which would not otherwise be available. Once we learn that eating toadstools causes stomach
pain, we then have open to us the opportunity to cause others to suffer stomach pain (by feeding them with
toadstools), to allow others (e.g. children) to be exposed to the risk of stomach pains (by allowing them to gather
toadstools without warning them of the possible effects), or to prevent others from incurring this risk. These
opportunities would not have been available without the knowledge; observation of natural processes producing pain
provides that knowledge. We know that rabies causes a terrible death. With this knowledge we have the possibility of
preventing such death (e.g. by controlling the entry of pet animals into Britain), or of negligently allowing it to occur or
even of deliberately causing it. Only with the knowledge of the effects of rabies are such possibilities open to us. That
knowledge is provided by observations of various people suffering subsequently to being bitten by dogs and other
animals with rabies in various circumstances. Or, again, how are humans to have the opportunity to stop future
generations contracting asbestosis, except through knowledge of what causes asbestosis? We can choose to obtain that
knowledge through laborious study of records which show that persons in contact with blue asbestos many years ago
have died from asbestosis thirty years later.

NATURAL EVIL AND KNOWLEDGE

193

Our study of nature may reveal processes with which we cannot interfere, but whose further consequences we may
learn to avoid by learning where and when they will occur. We may come to learn when comets will appear, volcanoes
erupt, or earthquakes strike, without (yet) being able to initiate or prevent these; but whose further consequences we
may be able to influence. Knowledge of when and where earthquakes are likely to occur gives us the opportunity
deliberately to cause, negligendy to risk, or, alternatively, intentionally to prevent suffering and death caused by
earthquakes, e.g. by taking the risk of building on areas subject to earthquake, or by making the effort to mobilize the
human race to avoid in future the consequences of a major earthquake.

The claim that a particular future C will cause an E will be justified by past observations, paradigmatically of past Cs
being followed by Es. Our a priori inductive criteria reveal that such a claim is better justified if it is based on many
recent purported observations of Cs in many different circumstances being followed invariably by Es. Observations
remote in time may have been misrecorded or have occurred under circumstances different in some way from those
holding in the present which affects the causal sequence in some crucial way. The more observations there are in
different circumstances, the better the evidence that the sequence of Cs being followed by Es is a genuine causal
sequence, not a mere occasional regularity. The observations might concern, not Cs and Es, but many very different
sequences which provide substantial but indirect evidence for a general scientific theory of which it is a remote
consequence that a C will be followed by an E. But the less similar is the evidence to the kind of phenomena predicted,
(because of the greater probability of alternative explanations of the evidence) the greater the doubt must be whether
the sophisticated scientific theory really works for Cs and Es; and that will require to be checked out by looking at
phenomena very similar to those predicted. Our knowledge of the future consequences of our actions is better justified
in so far as it comes from many recent observations in similar circumstances.

So sure knowledge that if I take frequent large doses of heroin I shall die must come from observations of frequent
heroin intake being followed by death. And many recent observations under different conditions provide the surest
knowledge possible. There could be a complicated scientific theory of which it was a remote

194

THE NECESSARY EVILS

consequence that heroin would have this effect. The theory would be confirmed by it being a simple theory which
yielded true predictions of the consequences of taking other, chemically similar drugs, perhaps tested sometimes on
animals rather than humans. The remoteness of the theory, and its never having been tested with respect to heroin on
humans, would, however, make its prediction about the effect of heroin on me much less well evidenced. And that
some drug causes pain is hardly likely to be even remotely evidenced, except via observation of other drugs causing
pain or other unpleasant sensations (so different are sensations from other things). Pain there must be which is
observed if we are to have knowledge of when our actions will cause pain.

The events by far the most important for the moral significance of actions which bring them about are mental events, i.
e. experiences of sentient beings. As noted earlier in this book, actions are paradigmatically good in so far as they
promote pleasurable and knowledge-deepening and friendship-deepening experiences; bad in so far as they promote
pain, ignorance, and poverty of imagination and understanding. Most sure knowledge of the experience to be caused
by some natural process is to be had through having experiences oneself of what followed from past occurrences of
the process. One knows best just what it feels like to be burnt by having been burnt oneself in the past. But the public
behaviour of others also produces strong evidence about their experiences. And if we have actually observed others
being burnt, we shall know quite a lot about what it feels like to be burnt. A person's knowledge is, however, less
securely based if the observations which support a theory are not their own, but ones known only through the
testimony of others; and that, of course, is the most usual case. My justification for believing that heroin causes death is
that everyone says that observers report (via television programmes and newspapers) that many who have taken large
doses of heroin have died quickly thereafter. This evidence, though good evidence, is always open to the possibility of
lying or exaggeration; or, where description of experiences is involved, lack of adequate vocabulary for the purpose.

So I condude this complicated discussion thus: if God is to allow us to acquire knowledge by learning from experience
and above all to allow us to choose whether to acquire knowledge at all or even to allow us to have very well-justified
knowledge of the consequences

NATURAL EVIL AND KNOWLEDGE

195

of our actions — knowledge which we need if we are to have a free and efficacious choice between good and bad — he
needs to provide natural evils occurring in regular ways in consequence of natural processes. Or rather, he needs to do
this if he is not to give us too evident an awareness of his presence.

Knowledge Provided by Animal Suffering

The suffering which provides knowledge is not confined to humans. The higher animals acquire knowledge by normal
induction, knowledge of where to obtain food, drink, and fellowship; and also knowledge of the causes of pain, loss of
health, and loss of life. Seeing the suffering, disease, and death of others in certain circumstances, they learn to avoid
those circumstances. And not merely do they observe and infer passively, but many animals actively seek knowledge:
they look for food, spy out the land for predators, put their feet gingerly on possibly unsafe surfaces, etc. The lower
animals of course avoid many situations and do many actions instinctively; but in those cases they cannot be said to be
doing the action or avoiding the situation through very well-justified knowledge of its consequences. We have noted
that acting in the light of such knowledge is a good thing even if the agent does not have free will; and so too, I suggest,
is actively searching for knowledge.

As I argued in Chapter 9, it is a great good that animals are not mere digestion machines with pleasurable sensations
attached to the digestive process; but that they struggle to get food, save themselves and their offspring from predators
and natural disasters, seek mates over days, and so on. But they can only do these things with some knowledge of the
consequences of their actions, and they could only acquire this by learning and seeking (as opposed to being born with
it) and it could only amount to very well-justified knowledge, if it is derived from experience of the actions of others, of
the unintended effects of their own actions, and of the effects of natural processes. Languageless animals could not
acquire knowledge by being told (by God or anyone else). They can acquire knowledge only by learning from
experience. Other animals must suffer if some animals are to learn to avoid suffering for themselves

196

THE NECESSARY EVILS

and their offspring. If deer are to learn how to help prevent their offspring from being caught in fires, some fawns have
to be caught in fires for the deer to see what happens. If gazelle are to learn to avoid being killed by tigers, they have at
least to have been mauled themselves or seen others mauled. Otherwise it will all reasonably seem a game. There will
not be any difference between playing ‘tig’ with tigers and playing ‘tig’ with other gazelles. And then animals will be
deprived of the possibility of serious and heroic actions.

The suffering of animals provides us, as well as themselves, with much knowledge; though since they are only
somewhat, not totally, like ourselves, the knowledge which the suffering provides is a less sure guide to what we would
suffer in certain circumstances than would be the suffering of humans; but still it is quite a good guide. Indeed, a great
deal of our knowledge of the disasters for humans which would follow some actions comes from study of the actual
disasters which have befallen animals. The bad states which have naturally befallen animals provide a huge reservoir of
information for humans to acquire knowledge of the choices open to them, a reservoir which we have often tapped:
seeing the fate of sheep, humans have learnt of the presence of dangerous tigers; seeing the cows sink into a bog, they
have learnt not to cross that bog, and so on. (And alas humans have for a long time chosen to increase their knowledge
of the effects of actions by deliberately doing things to animals which might cause them to suffer. They have
discovered the effects of drugs or surgery or unusual circumstances on humans by deliberately subjecting animals to
those drugs or surgery or circumstances. Before putting humans into space, we put animals into space and saw what
happened to them. Now I certainly do not think that humans had the right to do many such things. My only point here
is to illustrate the claim that animal suffering, however caused, often provides valuable knowledge for humans.)

And as regards very long-term consequences of changes of circumstances, environment, or climate, the story of animal
evolution provides our main information. Human history so far is too short to provide knowledge of the very long-
term consequences of our actions; and yet we are doing things which may have a considerable effect on the
constitution of the atmosphere (e.g. on whether there is still an ozone layer), on the balance of nature (e.g. on whether
there are many vertebrates on Earth other than humans), and on the climate. And we may discover how to make some
very big

NATURAL EVIL AND KNOWLEDGE

197

changes to the Earth and its surroundings, e.g. alter the Earth's magnetic field, drive the Earth nearer to the sun or
further away from it. And so on. We need information about the long-term effects of all these actions. There is a lot of
information to be gleaned from pre-hominid history on all these matters, for climate, magnetic field, and balance of
nature have changed often over the past 300 million years and if we learn more about their effects on animals we shall
avoid many disasters ourselves. But those effects must include suffering, in virtue of the similarities of animals to
ourselves. If there were not these similarities, the information would not show what would happen to us. To take but
one more and very strong example: biologists are beginning to acquire the power to cause much good or ill by
inducing various genetic mutations. Human history does not provide the data which will give them any knowledge of
the consequences of their actions. Their surest knowledge of those consequences will come from a study of the
evolutionary history of the consequences in animals of various naturally occurring mutations.

In addition to these detailed bits of information, the story of pre-human nature ‘red in tooth and claw’ already provides
one very general bit of information crucially relevant to our possible choices. For suppose that animals had come into
existence at the same time as humans, and always in situations where humans could save them from any suffering.
Naturally it would then seem a well-confirmed theory that (either through act of God or nature) suffering never
happens to animals except such as humans can prevent. So we would seem not to have the opportunity to do actions
which would cause suffering to present-day animals let alone later generations of animals of a subsequently
unpreventable kind, or the opportunity to prevent such suffering. We simply would not (and righdy would not)
seriously consider the possibility that some of our actions might have enormous and subsequendy unpreventable long-
term bad consequences. As evidence of this claim of mine, I point out that hardly anybody ever did consider such
matters before the nineteenth century. It is difficult to get back into a pre-Darwinian way of thought, but if you do, you
do not (and rightly do not) take seriously the possibility of our actions having long-term effects on nature. The story of
evolution tells us that the causation or prevention of long-term suffering is indeed within our power; such suffering
can happen because it has happened. The story of

198

THE NECESSARY EVILS

pre-human evolution reveals to humans just how much the subsequent fate of animals and humans is in our
hands — for it will depend on the environment which we form for them, and their genes, which we may cause to
mutate.

We may not know exactly when and where the past natural evils occurred, but the mere knowledge that suffering of a
certain type occurred to certain kinds of creatures under certain conditions provides us with very good reason to avoid
actions which may produce those conditions. Indeed all past evils of which we know provide knowledge of past events,
and, more strikingly, since all natural evils occur as a result of largely predictable natural processes (there are no kinds
of natural evil which occur in a totally random way), all such knowledge helps to build up knowledge of the natural
processes which we can utilize to produce or prevent future evils. All past and present human and animal natural evils
of which we know thus contribute to the widening of human choice when we learn about them. And (except at an
undesirable cost) we could not learn, and especially choose to learn, without them.

11 The Evils of Sin and Agnosticism

Now the serpent was more subtle than any beast of the field which the Lord God had made. And he said unto
the woman, Yea, hath God said, Ye shall not eat of any tree of the garden? And the woman said unto the
serpent, Of the fruit of the trees of the garden we may eat: but of the fruit of the tree which is in the midst of the
garden, God hath said, Ye shall not eat of it, neither shall ye touch it, lest ye die. And the serpent said unto the
woman, Ye shall not surely die: for God doth know that in the day ye eat thereof, then your eyes shall be opened,
and ye shall be as God, knowing good and evil. And when the woman saw that the tree was good for food, and
that it was a delight to the eyes, and that the tree was to be desired to make one wise, she took of the fruit
thereof, and did eat; and she gave also unto her husband with her, and he did eat. And the eyes of them both
were opened, and they knew that they were naked; and they sewed fig leaves together, and made themselves
aprons. And they heard the voice of the Lord God walking in the garden in the cool of the day: and the man and
his wife hid themselves from the presence of the Lord God amongst the trees of the garden.

(Genesis 3: 1-8)

So far I have been considering evils, moral and natural, which are bad states, whether or not there is a God. In this
chapter I pass to consider three states which occur or are bad only if there is a God: sin, punishment by God for sin,
and agnosticism. Sin, which is failure to perform our obligations to God, would not exist if there were no God; and
neither would punishment for sin. Agnosticism is bad only if there is a God to be known about. If there is a God, then
in containing these states, the world is a worse place than it would otherwise be; but, I shall be arguing, these very
states which make it worse make possible good states which could not otherwise occur.

200

THE NECESSARY EVILS

The Value of the Possibility of Sin

We saw in Chapter 6 that the existence of God makes an enormous difference to the moral quality of all intentional
actions. There are more obligatory and good actions, and more wrong and bad actions, than there would otherwise be.
All human acts of wronging others are acts of wronging God, that is sins. So too is any total failure to worship and
praise God and tell others about him. And failure to please God in many other ways, including the pursuit of our
vocation, becomes bad. And so, above all, does failure to seek the wonderful well-being (for ourselves and others) of
the Beatific Vision. Failure by the agnostic to attempt to find out whether or not there is a God is in any case (whether
or not there is in fact a God) also a moral evil. The reason why this is so is that whether there is a God or not makes
such a difference to what is our duty and what is good beyond duty to do; and it is good to put some effort into
finding out what our duties are, and it is good to find out what is good, beyond what duty requires. 141 But if there is a
God, then failure to investigate whether there is a God to whom duties are owed is not just the absence of a good but a
wrong, because there is now someone wronged by the failure — God.

The possibility of the occurrence (unprevented by God) of all this further moral evil has, however, the consequence
that we have far deeper responsibilities than we would otherwise have. We can give our lives to finding out whether
there is a God, and to worshipping and serving him if we find out that there is, or we can not bother. Our choices
affect our own destiny after this life and influence the destinies of others (for their destinies depend on their characters,
which we can help to form for good or refuse to do so). And in making all such choices we become responsible for
loving or hurting God. God, I argued in Chapter 6, will seek to be loved by those of his creatures who are capable of
understanding what he is like. While it is good that the love should involve a desire for God, it is good that it should
also involve freely chosen good actions of loving response to God. Love, I argued in Chapter 5, involves doing and
seeking to do things with and for the beloved. Doing and seeking to do things for God involves fulfilling our duties
and doing

141 On all this, see ag

ay Responsibility and Atonement (Clarendon Press, 1989), ch.

THE EVILS OF SIN AND AGNOSTICISM

201

and seeking to do supererogatorily good acts, including seeking to worship and interact with him. It is good that the
free choice of whether to make such a loving response to God should be a serious one: that we love God, not as
robots programmed to do so, but as free agents who have chosen to do so because they have seen how good God is
and resisted the temptation to pursue lesser goods. Hence without the possibility of moral evil resulting, not merely will
humans be deprived of the great good of a free choice between good and bad as such, they will be deprived of the
possibility of loving God in a very full way. God has every reason to provide for humans the possibility of such love
for himself. But, if someone is to have the opportunity to choose (despite contrary temptation) whether to give you his
love, he must also have the opportunity to hurt you instead. Hurting God means sinning. Vincent Briimmer describes
well the situation in which God must put himself if he is to give us that sort of opportunity to love him:

In creating human persons in order to love them, God necessarily assumes vulnerability in relation to them. In fact,
in this relation, he becomes even more vulnerable than we do, since he cannot count on the steadfastness of our
love the way we can count on his steadfastness ... If God did not grant us the ability to sin and cause affliction to
him and to one another, we would not have the kind of free and autonomous existence necessary to enter into a
relation of love with God and with one another . . . Far from contradicting the value which the free will defence
places upon the freedom and responsibility of human persons, the idea of a loving God necessarily entails it . 142

Although with the existence of God moral evil is worse by far than it would be otherwise, it serves the further
supremely good purpose of making possible a certain sort of very full love of God.

Vincent Briimmer, ‘Moral Sensibility and the Free Will Defence’, Neue Zeitschriftfiir systematische Theologie und Religionsphibsophie, 29 (1987), 86—100: 97 . The passage continues:
‘In this way we can see that the free will defence is based on the love of God rather than on the supposed intrinsic value of human freedom and responsibility.’ I have of
course argued that human freedom and responsibility do have great intrinsic value, but I agree with Briimmer that they have this very great further value as well. See also
Briimmer's valuable full development of the idea of love involving vulnerability to rejection in his The Model of Love (Cambridge University Press, 1993), esp. pt. IV .

202

THE NECESSARY EVILS

Punishment — Here and Hereafter

The Christian tradition has taught that God sometimes punishes wrongdoers for their actions by causing them to
suffer in this world — and so that some of this world's bad states are punishment for sin — and always punishes
wrongdoers who do not repent (or do not repent adequately) in the next world, either for a time or permanendy.

Wronging someone gives them (or someone else, such as the state, acting on their behalf) the right to deprive the
wrongdoer of some good that they would otherwise have had, or to impose on them some bad state which would not
otherwise be theirs . 143 That is their punishment. If you steal from me my watch, I have the right to take it back, and to
take more in return for the difficulty of getting it back and the trauma of loss. And if you cannot pay in kind, I surely
may choose in what form I take the compensation: I may deprive you of liberty, force you to work, force you to have
unpleasant experiences, up to some reasonable amount (in proportion to the harm you have caused me). If you
deprive my wife of life, I have the right (on her behalf), I suggest, to deprive you of your life. But rights are not duties,
and I do no one any wrong if I do not impose an equivalent punishment or any punishment at all; and it is sometimes
bad and more often not good for wrongdoers (or others on their behalf) to impose punishment. In particular the
imposition of pain or deprivation of life (from one who wishes to continue to live) is in itself a bad thing, and there
needs to be some good which it serves if it is to be good for the authority to impose it (although it has the right to do
so). For human imposers of punishment (parents and the state) there are often good utilitarian reasons for carrying out
the punishment (or some part of it): the traditional three reasons of prevention, deterrence, and reform. And all these
reasons are also reasons why it would be good for God sometimes to punish sinners on Earth.

God could punish me in some way which prevented me from sinning in that way again (he could cripple me so that I
could no longer pursue victims from whom I planned to steal); and preventing me from indulging the desire to pursue
and steal inevitably involves

On justifications of punishmer

e, more fully, Responsibility and Ato,

ch. 6.

THE EVILS OF SIN AND AGNOSTICISM

203

the suffering of frustrated desire. Such prevention of sinning is a good, although it is coupled with the deprivation of
the good of wide choice of action. Seeing me suffer harm can also deter others — God could allow others to see what
happened to me when I sinned (e.g. that the state imprisoned me). And earthly suffering is a spur to reform sinners
against God for the same reason as it is a spur to reform those whose wrongdoing is of a more secular kind (for the
reasons described on pp. 153-6). And God can provide these spurs to deter and reform without making his own
involvement in providing them too obvious (with a disadvantage which I shall describe later in this chapter).

The Christian tradition has also claimed that there is punishment after this life. It has affirmed fairly universally (until
the last two centuries, when a contrary minority position developed) that there is everlasting punishment for
impenitent sinners. And the Catholic and Orthodox branches of that tradition have held that there is also temporary
punishment for the inadequately penitent. Punishment of the former kind is punishment in ‘Hell’; punishment of the
latter kind the Catholic tradition has called punishment in ‘Purgatory’. The tradition has been that the everlasting
punishment consists of two parts: the poena damni, the penalty of the loss of God, the really heavy punishment; and also
a poena sensus , a suffering of some sort represented pictorially as the ‘flames of Hell’. There is, I think, no reason to take
the ‘flames’ too literally — the point is that there is suffering.

As regards the poena damni , I argued in Chapter 6 that our acts so mould our characters that firm and continued wrong
acts and lack of any regret for them will get us into the condition of incorrigibly rejecting the good. 144 It will only be a
bad thing that those in that condition should be deprived of the Beatific Vision if they still desire to have it. I pointed
out in Chapter 8 that one account of the state of the incorrigibly bad is that they have lost any desire for the good; and
so for them the poena damni will be in no way a bad thing. But if there remains in the incorrigibly bad some residual
desire for the good (and so, to the extent to which they know wherein it consists, for the Beatific Vision), clearly it is a
bad thing if it is not satisfied. The trouble is that the incorrigibly bad have, through

no everlasting punishment of the souls of the damned if they were able to change their will for a better will’ ( Summa contra Gentiles , 4

144 Aquir

rould be i

t 93. 2).

204

THE NECESSARY EVILS

their own considered choice over time, allowed themselves to develop stronger unalterable desires for states of affairs
incompatible with the desire for the good. They desire not to be penitent for their wrongdoing, not to be generous
with their lives, and to dwell on their own imagined greatness rather than worship God. A good God, I argued earlier,
will respect a considered choice of destiny. If their residual desire for the good remains through the choice of the
incorrigibly bad humans, God will respect that too; and the bad state that their desire for the good is unsatisfied will be
the known consequence of their own considered choice.

If, however, it was logically necessary that any human (or at any rate any human who once acquired such a desire) must
continue to have a residual desire for the good as long as he exists, then the only way in which God could eliminate the
bad state of it not being satisfied would be by eliminating the bad themselves; and maybe that is what God would do
with the incorrigibly bad — but surely only if that is what, after long consideration, they strongly desire. Either way,
whether or not the incorrigibly bad still have a residual desire for the good , poena damni will not serve any purposes of
prevention or reform, but if it is known to or suspected by others that that may be the fate of those who sin
continually, it can certainly deter those others.

It is good that God should provide humans with deterrents to sin, for the reason stated in Chapter 8. Humans are
weak. They need to be encouraged to do objectively good acts; and to start with, that may involve providing reasons
for doing such acts additional to the reason of their intrinsic goodness, tying the performance of such acts to lower-
grade self-centred rewards and punishments. To gain such rewards and avoid such punishment are of course good
reasons for acting, but there are better reasons. Once the human begins to get into the habit of doing the good acts,
doing them for the better reason becomes a serious possibility and threats are less needed. All parents and other
educators encourage good behaviour by threats and rewards as a preliminary to children learning to do good for its
own sake. And it would be good for God to do the same. 145 (Why make us so weak in the first place that we need
threats

‘Men are of three kinds: slaves, hirelings or sons. Slaves do not love the good, but refrain from evil out of fear of punishment; this ... is a good thing, but not fully in accord
with God's will. Hirelings love what is good and hate what is evil, out of hope of reward. But sons, being perfect, refrain from evil, not out of fear of punishment, but
because they hate evil violently; and they do what is good, not because they hope for reward, but because they consider it their duty. They love dispassion because it imitates
God and leads Him to dwell in them; through it they refrain from all evil, even if no punishment threatens them’ (St Peter of Damascus, ‘That Stillness is of Great Benefit’,
in G. E. H. Palmer, P. Sherrard, and K. Ware (ed. and trans.), Vhilokalia , , iii (Faber & Faber, 1984), 168.‘Fear of punishment hereafter and the suffering it engenders are
beneficial to all who are starting out on the spiritual way. Whoever imagines that he can make a start without such suffering and fear . . . thinks he can build in the air without
any foundations at all’ (St Symeon the New Theologian, ‘One Hundred and Fifty-Three Practical and Theological Texts’, in G. E. H. Palmer, P. Sherrard, and K. Ware (ed.
and trans.), Philokalia, iv (Faber & Faber, 1995), 37-8.

THE EVILS OF SIN AND AGNOSTICISM

205

and rewards? To allow us, as I argued in Chapters 7 and 8, by our own choices to form our characters, and to allow
others to help us to do so. That can only happen if we do not have a firm character to start with.)

Now the deterrence could be provided either by showing what happens to others (showing others being punished), or
by a vow to impose punishment which the hearers believe would be fulfilled. As we noted, humans can be deterred
from doing actions of some kind by seeing others punished (or at any rate suffering) on Earth when they do actions of
that kind. But there is obvious advantage in a mode of helping to deter some which does not involve the actual
punishment of others, and hence the advantage of the second method of deterrence. So God has reason to vow to
punish in the next life (temporarily or permanendy) those who sin in this life. Most major theistic religions, including
Christianity, have taught that God does so vow We saw in Chapter 5 that when an authority vows to punish those who
do certain acts, in order to deter them from so acting, it is good that it should execute that punishment if the acts are
done. And when the punishment consists simply in the loss of the Beatific Vision, which it is not good that those who
have corrupted themselves beyond correction should have anyway, there is a point in God vowing publicly so to
punish them — to be weighed against a major consideration to be adduced later in this chapter.

But why a further punishment {poena sensus) as well? That surely is a bad thing. But the threat of poena sensus will
(because of the strong desire to avoid it, which is (see Chapter 4) the essential part of suffering) be a more powerful
deterrent to the hard-hearted who have litde love for God, than the threat of poeni damni. The former threat may — and
through the centuries so often has — started the hard-hearted on the road to sanctity. And there is some good, apart

206

THE NECESSARY EVILS

from the fulfilment of a vow, in allowing poena sensus to occur. For those who reject the good will probably be left with
substantial desires for what is wrong, not merely for the bad. They will want to deceive and hurt others. But while there
is good in allowing them sometimes to satisfy those desires before their evil character is formed (i.e. when there is
some possibility that they will resist those desires), it would be bad and indeed wrong of God to allow them to satisfy
their desires everlastingly. Yet God would violate the Principle of Honesty if he allowed them permanendy to believe
that their desires were being satisfied when they were not. So inevitably if such people continue to exist, they will have
frustrated desires (even if those desires do not include a residual desire for the good). That suffering can hurt quite as
much as pain which has merely physical causes. Maybe again it would be best if God simply eliminated such people;
and if that is what they firmly desire, maybe he will do so. Yet if they go on existing with the character they have firmly
chosen to have, they will suffer. And surely God would not choose to cause them not to exist, against their wishes.

Maybe some of those who have firmly rejected the good are left only with desires for what is bad, including everlasting
enjoyment of the trivial (quite out of proportion to its worth) but not for what is wrong. The mere having of bad
desires, satisfied or not, is a bad state for their possessor; and constitutes itself a punishment but not suffering . 146 And
maybe God will allow such desires to be fulfilled. Such creatures could live for ever without poena sensus in the sense of
suffering (George Bernard Shaw's vision of Hell in Man and Superman is a life of people following trivial pursuits for
ever). It is not at all obvious, however, that the desire for everlasting Veil-being’ of someone whose idea of well-being
was limited to very trivial goods is a desire for well-being and so one which God would not leave frustrated. A desire
for chocolate tomorrow is a good desire, but a desire for nothing but chocolate for ever seems to be a desire so out of
proportion to the worth of what is desired as to be a bad desire which it would not be good for God to satisfy.
However that may be, for those with wrong desires fixed incorrigibly there will inevitably be suffering. Yet the
Christian tradition has been far from unanimous over the kinds and lengths of punishment

saw it as an appropriate punishment for the wicked that they should ‘serve other gods day and night’ (Jer. 16: 13), i.e. ones less w

Jeremiah

worthy of s

THE EVILS OF SIN AND AGNOSTICISM

207

of the bad; 147 and I side with some of its more moderate exponents in holding (in effect) that God would not impose
permanent intense suffering on wrongdoers who do not desire to go on existing.

What of temporary punishment for the inadequately penitent? The inadequacy of their penitence shows a mixture of
desires good and bad. Penitence involves a resolution to reform, and that can only be made by someone who has some
desire to reform or sees it as a good to reform. But if the penitence is half-hearted, the penitent has allowed other
desires to influence him. Such a person has not firmly chosen what character he shall have; and for reasons yet to be
discussed the overwhelming presence of God would inhibit such free choice of character. Hence the need for a further
temporary poena damni to give the opportunity for reform. (Maybe, as in the Catholic version, Purgatory is so arranged
that its ‘inhabitants’ cannot, in virtue of the good desires they have formed or kept intact on Earth, ever finally lose the
Vision of God. In that case their choice will be simply between staying put or finally and firmly choosing the good.)
This intermediate state might involve suffering, for the same reason as the permanent state, through the frustration of
wrong desires. The punishment of the intermediate state would serve the purposes not only of deterrence (if others on
Earth know of it) and prevention (of satisfaction of wrong desires), but above all of reform; i.e. it would provide the
opportunity for the sinner to choose to reform.

The Religious Value of Worldly Failure

I argued in Chapter 8 that it is good that there should be desires which provide natural incentives to do good actions
and deterrents from doing bad ones — temporarily, when we are ignorant or lazy or start to go wrong. A good God,
like a good parent, would not let us make a mess of our lives without providing temporary encouragement to do good
and discouragement from doing bad. My concern there was mainly with the biologically produced desires to do

147 For a fuller account of the differences in this respect in the tradition, and justification for my siding with the more moderate part of it, see 'Responsibility and Atonement , ch. 12.

208

THE NECESSARY EVILS

actions for our mundane well-being. But it is good also for God to provide (whether or not in the form of a
punishment here or hereafter) temporary spurs and deterrents in regard to the action of seeking God, and to the other
God-directed actions of worship, petition, and evangelism which it is good that we pursue. If we neglect such actions
because we yield to mundane desires for food and drink, fame and fortune, it is good that these should be frustrated
(temporarily); and that means suffering. The existence of God has the consequence that frustration of desire provides
an opportunity to develop better the desires whose fulfilment lies outside this world. You can resist that opportunity by
indulging in such worldly pleasures as remain open to you or by self-pity. But if past pleasures yield pain, that at any
rate gives you an opportunity to rethink your goals. The Israelites were ‘bitten’, says the Wisdom of Solomon
addressing God, ‘to put them in remembrance of thine oracles ’. 148 And as Eleonore Stump has written, echoing so
many others: ‘Natural evil — the pain of disease, the intermittent and unpredictable destruction of natural disasters, the
decay of old age, the imminence of death — takes away a person's satisfaction with himself. It tends to humble him,
show him his frailty, make him reflect on the transience of temporal goods, and turn his affections towards other-
worldly things, away from the things of this world .’ 149

True, I am suggesting both that it is good that people have strong desires for worldly well-being (food, drink, health,
success in work and love, a happy domestic life, and so on) and also that it is good that such desires are sometimes
unsatisfied, so that we suffer in consequence. The satisfaction of the former desires is good. But we can let them
become too important to us. Peter of Blois wrote:

You could say that tribulations are unnecessary to induce this remembrance of him, for God by conferring benefits
gives us sufficient warning, as Augustine says, that God's kindnesses are nothing other than reminders that we
should proceed to him. So it should be enough for God to convey warnings by conferring kindnesses, for such
warnings befit God more than those delivered with canings. To this a possible reply is that though kindnesses recall
you to acknowledge him, on occasion an uncontrolled love for

Wisd. 16: 11.

Eleonore Stump, ‘The Problem of Evil’, Faith and Philosophy, 2 (1985), 392-423: 409 . See also C. S. Lewis, The Problem of Pain (Fontana Books, 1957), ch. 6 . Pain ‘shatters
the illusion that all is well’ and then ‘the illusion that what we have, whether good or bad in itself, is our own and enough for us’ (p. 83).

THE EVILS OF SIN AND AGNOSTICISM

209

those very kindnesses holds us fast, and the Creator, the highest unchangeable God who bestows eternal blessings,
is then forgotten. 150

If we are to have a serious choice of Heaven-seeking, we must have, to start with, mundane desires stronger and
Heaven-seeking desires weaker than in proportion to their worth. But if we yield too much to worldly desires, it is then
good that we are encouraged to pull ourselves out of submission to the former by some of them proving unsatisfiable.
Failure in love or work lets us rethink our priorities. It is not necessary that we learn Heaven-seeking by suffering; but
there is a good, if we fail to learn it by any other route, that this final spur be available.

The Value of Agnosticism

We saw in Chapter 6 that if there is a God, among bad states is the fact that (even on Earth) so many who long for
God have no vision of God (not even a limited one), doubt his existence, and often believe that he does not exist. I call
this non-awareness of God, in all its various forms, agnosticism. If there are good arguments for the existence of God
(as I think that there are), then some human ignorance of God may be moral evil. Humans may not have bothered to
consider these arguments seriously, may not have told each other about them, and may have refused to allow
themselves to feel the force of the arguments because in consequence they would see the obligation to lead a different
sort of life. The same goes if there is an experience of God to be had for those who, having heard some preacher tell
them what God is like, begin to pray to him. Humans may not have bothered to preach the Gospel message or
respond to it. And above all, it would be a moral evil if someone had an experience of God and then tried to persuade
themselves that it did not occur.

In a situation where a belief which is in some way unwelcome is only of moderate strength, there will, I noted earlier,
always be the temptation to self-deception, to get ourselves not to believe by

Commentary on the book of Job . He sets out Twelve Advantages of Tribulation’ (PL 207.989). English translation in J. Walsh and P. G. Walsh (eds.). Divine Providence and Human
Suffering (Michael Glazier, 1985), 141-62: 149 .

210

THE NECESSARY EVILS

irrational means. If there is some doubt about whether there is a God, although on balance evidence or experience
suggests that there is, we may choose (subconsciously) to set about deceiving ourselves into supposing that the
evidence is misleading or the apparent awareness of God illusory. But self-deception is never easy. We cannot just
choose there and then to believe things or not to believe things (see p. 55). It takes time and effort to force oneself to
believe something contrary to the way the evidence seems to point. And the more something stares you in the face, the
harder it is to persuade yourself that it is not really there. No doubt some theists become atheists by this route (as no
doubt some atheists become theists, or theists or atheists keep themselves theistic or atheistic also by self-deception).
But clearly not all human ignorance of God is due to refusal to teach or preach, or to listen to teaching or preaching, or
to self-deception. Some are unmoved by preaching and in all honesty cannot see the force of such arguments as they
have heard. 151 Some who have prayed have not come to have any awareness at all of God's presence. So much
agnosticism is a natural evil.

If there is a God who is our loving creator, why does he not make his presence far better known in the world? Fathers
who absent themselves too much from their children are rightly judged less than adequately loving. God's failure to
make himself known is surely, an objector will say, 152 in view of God's supposed perfect goodness, evidence against his
existence.

My answer to this objection is that there is, similar to the incompatibility between the good of a free choice between
good and bad and the good of a spontaneous inclination to do good, an incompatibility between the great good of my
having a deep awareness of the

One cause of human ignorance of God is the belief of so many that moral and natural evil conclusively disprove or at least render improbable the existence of God. And so
theodicy needs to explain why (if this is not so) God should allow many humans to think that it is. This chapter thus seeks to provide an explanation of why this book
(among innumerable other means of presenting theodicy) needs to be written.

This objection has recently been presented very thoroughly in a book devoted solely to the present topic: Divine Hiddenness and Human Reason, by John Schellenberg (Cornell
University Press, 1993) . Schellenberg's positive claim is that a perfectly good God would provide ‘probabilifying evidence’ (p. 35) of his existence (whether by experience of
God or the availability of public objective argument) for all capable of being aware of God, £ at all times’ (p. 25); and that since we do not all always have such evidence, there
is no God. This chapter constitutes my answer to Schellenberg's fine book.

THE EVILS OF SIN AND AGNOSTICISM

211

presence of God, and other good states of myself and others. The first of these latter is just the libertarian free choice
between good and bad which I have been considering in earlier chapters. This awareness is incompatible with such a
choice, given that I have either a strong desire to be liked by good persons (and especially any on whom I depend for
my existence), stronger than any contrary bad inclination; or a strong desire for my own future well-being combined
with a strong belief that it is quite likely that a God would not provide a good afterlife for bad people.

Before coming to detailed argument to show these incompatibilities, I draw attention to a general point. Two desires
for different goals, or a desire for some goal and a moral belief about the goodness of some action, will only be in
conflict with each other, only prescribe different actions in a given situation, given beliefs about how to secure the
goals. My belief that I ought not to lie will only be in conflict with my desire for money if I believe that by telling some
lie, I can acquire money. If I believe that my lie will be discovered to be such and not lead to my enrichment, there will
be no conflict. Or again, my desire to eat will only be in conflict with my desire to be thin if I believe that eating will
make me fat. If the belief which introduces a conflict between a desire and a moral belief about the goodness of some
action is not very strong, if it is only a balance-of-probability belief, then the influence of the desire upon my choice will
be correspondingly weak. My desire for money or love or knowledge may be strong; but if my belief that the only route
known to me is only somewhat more likely than not to achieve my goal, the influence of the desire on my conduct will
be correspondingly reduced.

I begin by considering the effect of the desire to be liked, which, as we saw in Chapter 4, it is good for any person to
have. It is a good for me to like (i.e. to have the desire or inclination) to be liked, to like to be thought well of by others,
and to have them seek my company and respect my opinions and achievements. To like and to like to be liked are
essential elements of friendship. And, as we saw in Chapter 6, friendship with the good, and above all with the
perfecdy good perfecdy wise all-powerful source of my existence, would be an enormous good. There would be
something deeply wrong with someone who did not mind whether such a God liked him much or not. And yet how
could such a God like him very much if he did and was inclined to do wrong actions? For God loves

212

THE NECESSARY EVILS

the good and hates evil, and while he may still love us because he has made us and we still have some potential for
doing good, he will not love us for our wrong actions.

Now if I acquire a deep awareness of the presence of God, I will then become deeply aware that if I do bad, and
especially wrong actions, the all-good creator will strongly disapprove. Hence if I have the proper desire to be liked, I
will have a strong inclination not to do wrong; and unless that is overborne by some even stronger desire to do wrong,
there will be a balance of desire against choosing wrong and so no overall temptation to resist reason. I will inevitably
do the good.

The desire to be liked may be of various strengths, as may the desire to do wrong, and the belief that there is a God.
But if the good desire is stronger than the bad one and I have a deep awareness of the presence of God (i.e. such that
God's existence is not open to question), then the balance of inclination will be to the good and there will be no free
choice between good and wrong. We will be in the situation of the child in the nursery who knows that mother is
looking in at the door, and for whom, in view of the child's desire for mother's approval, the temptation to
wrongdoing is simply overborne. We need ‘epistemic distance’ in order to have free choice between good and evil.

The only way in which a strong awareness of the presence of God will leave open the possibility of free choice between
good and evil will be if the desire for divine approval is weaker than the desire to do wrong. If God makes us naturally
malicious enough, keen to hurt and deceive others with no natural affection for them, then the choice may remain
open. If our desire to hurt others is strong enough, we will have a serious temptation to yield to the desire even if we
believe that God is watching us, so long as we care only moderately about what he thinks about us. If we feel a very
strong desire to hit our brother, the belief that mother is looking in at the nursery door will be insufficient to ensure
our good behaviour, unless we want to please mother fairly strongly . 153 But of course just as it is good that we desire
the love of God, so it is bad if we are naturally malicious and lack natural affection. Yet it is not logically possible

And maybe the situation of Satan, if he was strongly aware of the presence of God and yet wished to rule the world, was like that when he was first created by God: he was
subject to such a strong passion of pride and such a weak desire for divine approval that the possibility remained open that he would yield to pride.

THE EVILS OF SIN AND AGNOSTICISM

213

that God give us both a strong awareness of his presence and a free choice between good and evil at the same time as
giving us a strong desire for his love and some natural affection for our fellows. These latter are great goods, and so
again the logical straitjacket. God can only give us certain goods if he does not give us others.

God could give us a much more moderate awareness of his presence, ambiguous experiences but ones best interpreted
as experiences of God, or arguments to show that on a slight balance of probability there is a God. Then a strong
desire for his love will, despite its strength, be less influential on our conduct. For it will only be on balance probable
that we will satisfy that desire by acting in the right way, whereas it may be much more nearly certain that we will satisfy
our bad desires (e.g. to hurt) by some route (e.g. by hitting the hated one). If there is a doubt whether mother is looking
in at the nursery door, we may get away with hurting our brother without displeasing her.

Yet even given only a moderate belief that there is a God, the possibility of a free choice between right and wrong will
only exist given a certain ratio of strength between the desire to please God and the desire to do wrong. Even if the
influence of the former is modified by doubt about the existence of God, the former cannot be too much stronger
than the latter if there is to be that balance of desire for wrong action over right which alone makes possible the very
serious free choice between right and wrong. Even God cannot give us that choice if he gives us fairly strong natural
desires for good (including the desire to be liked by the good) and shows us, even only on a balance of probability, that
he exists. But the more uncertainty there is about the existence of God, the more it is possible for us to be naturally
good people who still have a free choice between right and wrong.

This point is strengthened when we take into account the other relevant desire — the desire for our own future well-
being, another desire which it is very good that we should have (as I noted in Chapter 4). If we have no belief that God
will allocate a fate to us in the afterlife dependent on the way we live in this life, then of course that desire will not
interact with the belief that there is a God to affect our conduct. But a belief that if there is a God he will give us future
well-being only if we act well (even if not a very strong belief) will influence the desire for our future well-being so that
it combines with the desire to be liked by God to produce a greater

214

THE NECESSARY EVILS

balance of desire in favour of the good. Our malicious desires will need to be stronger, or our desire for our future
well-being weaker, or our belief that there is a God even weaker, if there is to be that balance of desire in favour of bad
requisite for serious free choice. If any God there may be does allocate fates in the afterlife dependent on the way we
live in this one, then it is good that we should believe this. And then, other things being equal, the belief that there is
indeed a God will deprive us of that serious choice between good and bad — unless God gives us weak desires for the
good and strong desires for the bad. Again, we have the logical straitjacket; if God gives us some good things, he
cannot give us other good things.

The belief that there is a God inevitably gives some degree of probability to the belief that there is for all who desire
their own future well-being some life after death. For God will regard it as a good thing that there be no unsatisfied
desires. But, as I have commented earlier, humans by their choices over life gradually mould their character. Those
who do bad and wrong persistendy lose their sensitivity to the good; and so form an incorrigibly bad character. Such
people would cease to desire their future well-being (in my objective sense of ‘well-being’), and so for the reasons given
earlier in this chapter there would seem litde good in keeping them in existence. They might desire to continue to exist,
but their other desires being for the bad, they will eventually be frustrated, and so for the reasons given earlier in this
chapter there would seem litde good in keeping such people alive; and, if God did keep them alive, frustration of desire
would make them unhappy. I conclude that anyone who believes that there is a God and reflects on the fact that bad
acts destroy character will realize that they risk their eternal well-being by performing them.

While a priori considerations about what God might be expected to do with the bad in the afterlife are only of
moderate force, virtually all the major religions which believe that God has revealed truths believe, as we have seen,
that he has revealed that those who have done wrong will be punished in the afterlife. Any reason for believing that
there is a God is some reason to believe that he would tell humans (who so obviously need guidance ) 154 wherein
consists human good, and so is some reason for believing that what is the

iy Revelation (Clarendc

Press, 1992), ch. 5 .

THE EVILS OF SIN AND AGNOSTICISM

215

common content of virtually all theistic religions is true. So on both a priori and a posteriori grounds, any reflective
person will conclude that the more reason there is for believing that there is a God, the more reason there is for
believing that if we continually do wrong, we shall not have a good afterlife. The desire for our own future well-being
will thus make it far easier for us to choose the good. The more God manifesdy encourages us to do good, the less
free choice we have of whether to do it.

It will be possible to combine a moderately strong belief that there is a God, with bad desires and some significant
desires to please God and to have a good future, and also a free will to choose between good and bad, when the choice
is quick and instinctive and not deliberate, and the bad not so obviously bad. As I noted in Chapter 6, libertarian free
will is less in control in such a situation; and hence venial but not mortal sin is possible. That, I suspect, was the
situation of St Paul when he wrote ‘the evil which I would not, that I practice’. 155 But it is good that we shall have the
big serious choices between the good and the wrong, which we can take seriously and deliberately after reflection. As I
have argued, a generous God might well seek to give us the opportunity to reject all that he stands for, i.e. to commit
mortal sin. And of course doing mildly bad acts which we know are bad makes it easier to do serious bad acts.
Persistent venial sin opens up the possibility of mortal sin, a danger of which St Paul was well aware: T buffet my body
and bring it into bondage: lest by any means, after that I have preached to others, I should myself be rejected.’ 156 But St
Paul is not, as he writes, in the position of being able to reject God and the good for which he stands; and it would be
very wrong if he allowed himself to get into that position. But someone who finds himself in that position (not through
his own choice) does have the advantage of being able to make the very serious choices which St Paul cannot make.

In summary of the argument so far: the greater our conviction of the existence of God, if we have one or other of two
very good desires (the desire to be liked by the good, and the desire for our future well-being), the less it will be open
to us to make serious choices, including the choice of our ultimate future.

There are significant biblical passages which bring out the important role of ignorance of God in making possible
serious, and

Rom. 7: 19.
I Cor. 9: 27.

216

THE NECESSARY EVILS

especially very serious, free choice; and they include the Bible's account of that famous first very serious free choice
cited at the beginning of this chapter. The serpent persuaded Eve to take the apple on the grounds that ‘God’ was
weak, i.e. not God: for those who eat the apple would be as powerful as God — something which Adam still believed
after eating the apple, because he supposed that by hiding himself he could escape God. And the more general point
that serious human choices are often made with inadequate knowledge of their divine implications is also very much a
New Testament one. In the parable of the sheep and the goats at the Last Judgement, as related by St Matthew , 157 the
good are rewarded for feeding the hungry, visiting the sick, etc., in ignorance that the beneficiary of their acts was God
himself. (Tn so far as ye did it unto one of these my brethren, even these least, ye did it unto me,’ said the king.) And
the bad are punished for their failures to act in similar ignorance. Their amazement at the punishment (‘when saw we
thee an hungred?’ etc.) implies that if they had known who the potential beneficiary was, they would have fed him.
Ignorance of God is a precondition here for the sheep making different choices from the goats. ‘Whatever you are
doing, remember that God sees all your thoughts, and then you will never sin,’ wrote St Isaiah the Solitary . 158

The belief that there is a God does not merely (given other good things) deprive us of a certain sort of free choice; it
deprives us also of the possibility of doing various different actions, and so of the choice of so doing — whether free or
not, contrary to strong or weak temptation. The first of these is being able to find out for ourselves whether or not
there is a God. If, as I have been arguing throughout, doing things for ourselves, i.e. intentional action which
endeavours to make a significant difference to our condition and that of others, is a good thing, to have a free choice
of whether to bother to try to find out whether there is a God will be a great good. And a greater good will be to find
the answer. All that follows from the supreme good of knowledge of the ultimate truth about the Universe and so of
whether there is a God. But to seek, let alone to find, knowledge, we need ignorance to start with. And in this situation
of ignorance there will be the possibility of seeking and

Matt. 25: 31^6.

‘On Guarding the Intellect

t’, in G. E. H. Palmer, P. Sherrard, and K. Ware (ed. and trans.), Vhilokalia , i (Faber & Faber, 1979), 28 .

THE EVILS OF SIN AND AGNOSTICISM

217

obtaining not merely theoretical knowledge but the deep friendship with God which is such a good thing.

Joseph Buder emphasized the value of investigation in discovering the content of revelation and of uncertainty about
it:

If a prince desire to exercise or in any sense prove, the understanding or loyalty of a servant, he would not always
give his orders in such a plain manner . . . Ignorance and doubt afford scope for probation in all senses . . . Men's
moral probation may also be, whether they will take due care to inform themselves by impartial consideration, and
afterwards whether they will act as the case requires upon the evidence which they have, however doubtful. 159

Any goal which is good for humans to pursue is, I argued in Chapter 5, especially good if they pursue it cooperatively.
In the pursuit of this very great good, the knowledge of God, it is good that humans cooperate, cooperate in
investigation when none have found the great good; and when some have found it and others have not, that those who
have found should try to help those who have not, to find it. What a good thing it is that those who know should be
able to teach those who do not — in all matters, but above all in this most important of matters! 160 Cooperative inquiry
into the existence (and nature and activity) of God will not only take the form of ‘centres for the philosophy of religion’
(though it will include that). It will also take the form of parish discussion groups, and prayer meetings, for seeing
something about God is not confined to the learned, and requires asking his help (if he is there). And teaching the
ignorant always involves a lot of publicity to persuade them that they need to learn and that certain teachers are the
best people to teach them. Hence the goodness of evangelism. But of course cooperative inquiry into a matter over
time presupposes a general previous ignorance in the population until that time. Only those who do not want to learn
can be persuaded to do so, and only the ignorant can be evangelized. If God is to give us the great good of helping
each other to find him, he must make a lot of us ignorant of him over a long period. And, given the way generations
inevitably depend largely on previous generations for their knowledge

159 J. Butler, The Analog ofRe/igon, pt. 11, ch. 6 (George Bell, 1902), 272-3.

160 The ignorance on the part of the Jews of God's activity in Christ which evoked St Paul's desire and evangelistic effort to bring all to Christ led to that most
expressions: T could wish that I myself were anathema from Christ for my brethren's sake, my kinsmen according to the flesh’ (Rom. 9: 3).

: generous of all his

218

THE NECESSARY EVILS

(and what a good thing it is that parents are of such use!), that means that there must be generations and cultures
ignorant of God, if the human race are to be of immense use to each other in this way . 161

The Advantages of Human Death

I argued earlier that the belief that there is a God inevitably gives some degree of probability to the belief that there is
(at least for the good) life after death, and that that makes it easier for us to do good; temptations to do bad are not so
acute, and so we do not have the opportunity to stand up for the good so heroically. While this is so with respect to all
good and bad actions, it is also the case, as I shall show below, that certain kinds of heroic and wicked action become
unavailable to us to the extent to which we believe that there is life after death.

It will be useful at this stage to summarize the advantages (some of which I have mentioned earlier) of a world in
which humans live a short, finite life, culminating in a natural death, but which they can cause to happen earlier (by
killing each other or themselves); and then point out that in two respects these advantages are lessened to the extent
that humans believe that death is not the end. I wrote earlier that death is only a bad thing to the extent to which it
involves frustration of desire — i.e. the desire to live; and while not all humans always desire to go on living under their
this-worldly conditions of existence, they almost all wish to go on living under some conditions of existence. Hence if
death was really the end, it would be a bad thing for this reason as well as, of course, in so far as it involves the grief of
others. But the fact of death does confer on us great advantages.

The first advantage of death is that it allows agents to do or to refrain from doing to each other a certain harm (of a
qualitatively different kind to other harms) — to deprive of existence. In this respect humans would be like God, able to
destroy. And God would indeed have trusted humans enormously in giving them this power.

The second advantage of mortality is to give us the possibility of

161 See Additional Note 7 .

THE EVILS OF SIN AND AGNOSTICISM

219

supreme self-sacrifice and courage in the face of absolute disaster. The ultimate sacrifice is the sacrifice of oneself, and
that would not be possible in a world without death. (‘Greater love hath no man than this, that a man lay down his life
for his friends,’ said Christ. 162 ) If we are immortal, supreme generosity is impossible. So too is cheerfulness and patience
in the face of absolute disaster. For in a world without death the alternatives would always involve continuance of life
and presumably too therefore the possibility that others would rescue one from one's misfortunes. There will be no
absolute disaster to be faced with cheerfulness and patience.

Thirdly, a world with natural death is a world in which an agent's own contribution has a significance to it because it is
irreversible by the agent. If I spend all my seventy years doing harm, there is no time left for me to undo it. But if I live
for ever, then whatever harm I do, I can always undo it. It is good that what people do should matter, and their actions
matter more if they have only a limited time in which to reverse them.

Fourthly, a world with birth but without natural death would be a world in which the young would never have a free
hand. And birth is an evident good. It is good that some should have the responsibility for the existence and growth to
maturity of others. Yet without death those others would always be inhibited by the experience and influence of the
aged — given the obvious goodness that, if the aged do live for ever, they continue to possess old knowledge, acquire
new knowledge, and are respected for their knowledge.

The greatest value of death, however, seems to me to lie in a fifth consideration, which is in a way opposite to my
second one. I wrote earlier of the great value which lies in agents having the power to harm each other. Only agents
who can do this have real responsibility. Yet it may seem, despite the arguments which I gave earlier, unfair that agents
should be too much subject to other agents. Clearly for the sake of the potential sufferer, there must be a limit to the
suffering which one agent can inflict on another. It would, I believe we would all judge, be morally wrong for a very
powerful being to give limitless power to one agent to hurt another. Giving to agents the power to kill is giving vast
power of a qualitatively different kind from other power: but it involves the end of experience.

John 15: 13.

220

THE NECESSARY EVILS

It is very different from a power to produce endless suffering. Clearly the parent analogy suggests that it would be
morally wrong to give limidess power to cause suffering. A parent, believing that an elder son ought to have
responsibility, may give him power for good or ill over the younger son. But a good parent will intervene eventually if
the younger son suffers too much — for the sake of the younger son. A God who did not put a limit to the amount of
suffering which a creature can suffer (for any good cause, including that of the responsibility of agents) would not be a
good God. There need to be limits to the intensity of suffering and to the period of suffering. A natural death after a
certain small finite number of years provides the limit to the period of suffering; and I shall claim in Chapter 12 that
God does not have the right to entrust humans to the care of other humans who may hurt them for more than a short
finite period.

Now the Christian tradition teaches that although death is not the end, it does involve God removing humans from the
web of causes and responsibilities provided by life on this Earth. The fourth and fifth and (to an extent) the third
advantages of death are therefore preserved. The young on Earth are no longer inhibited by the old, nor can any one
human continue to harm another over the gulf provided by death. Our removal from this Earth takes us away from
being able to undo harm done to things and other humans on Earth, but we may be able to undo the harm done by
our actions towards God. But the first and second advantages are somewhat lessened. To kill another is to deprive him
or her of life — on this Earth with its web of causes and responsibilities — but not of existence; it is rather to force the
one killed (unprepared) into the nearer presence of God. That is indeed a significant action, but not as significant as
depriving someone of existence for ever. And in consequence, self-sacrifice and cheerfulness in the face of death do
not have quite the absolute character they would otherwise possess.

Yet while our responsibilities (what we can effect by our choices) will be less than they are, if in fact God gives us a life
after death, to the extent to which we believe that this will not happen, the seriousness of our choices will be
unimpeded. Ignorance of whether death is the end means that our choices of whether to kill another or sacrifice
ourself become very serious indeed, for we suspect that they will involve deprivation of existence for ever; and the
temptation to avoid self-sacrifice becomes enormous. I see a close and

THE EVILS OF SIN AND AGNOSTICISM

221

rational connection between the deep unwillingness today of Western nations to send troops to fight for justice in
distant lands, and the decline in the West of religious faith. Death is a grimmer thing than it used to be.

The Incarnation and the Atonement

These are two doctrines which are above all the hallmark of Christianity, which each have the consequence that our sin
which is so evil makes possible a very great good.

The doctrine of the Atonement is that the life and death of Christ on the cross was an atonement for our sins. We owe
God much, who has so generously given us our lives; and if we waste those lives or use them to do wrong, we must
apologize and offer reparation: to fail to do so would be to fail to take our wrongdoing seriously. But we are in no
position to make reparation to God, since we owe him so much anyway. So God made available to us a life which he
lived himself in Christ, which we can offer back to God as reparation; in baptism and the Eucharist and other church
ceremonies we can plead in atonement for our sins, the sacrifice of Christ's life and death . 163

Now the atoning life and death of God in Christ were marvellous things; the goodness of generous service was a
major theme of my Part II, and that God himself should live such a life so that we can take our wrongdoing against
him seriously is wonderful indeed. It cannot, however, be a good thing that we do wrong, simply in order that God
may make an atonement and forgive us. For wrongdoing would not be wrongdoing (and so could not be atoned for) if
it were good that it be done. If wrongdoing were overall a good thing, the answer to St Paul's question ‘shall we
continue in sin, that grace may abound?’ would be ‘Yes’ instead of ‘God forbid ’. 164 Still, the difference which it makes is
this: while the possibility of its misuse provides a reason for God not to create creatures with significant

This is a very brief account of what seems to me the most plausible way of spelling out the doctrine of the Atonement. But further spelling-out is clearly needed. (For that,
see my Responsibility and Atonement , esp. ch. 10.) The point of this section, however, that the life and death of Christ, in atoning for our sins, lessens the force of the reason
which God had for not creating creatures with significant freedom, remains also on most other accounts of the Atonement.

Rom. 6: 1-2.

222

THE NECESSARY EVILS

freedom, the possibility of such compensation for misuse reduces, if not balances, the force of that reason. There is
great truth, though maybe not as much as the writer of the Exultet supposed, in ‘O Felix Culpa, quae talem ac tantum
meruit habere redemptorem .’ 165 That he was willing to make such an atoning sacrifice meant that God had more
justification than he would have otherwise for taking the risk of allowing us to do wrong.

The doctrine of the Incarnation (that God was incarnate in Christ) has its relevance to human suffering, quite apart
from its role in atonement. For the argument of this book is that it would have been good for God to subject humans
to much that is bad, especially suffering, for the sake of great goods, and I have been seeking to show what these
goods are and how suffering is needed for their attainment. We humans sometimes righdy subject our own children to
suffering for the sake of some greater good (to themselves or others), e.g. make them eat a plain diet or take some
special exercise for the sake of their health, or make them attend a ‘difficult’ neighbourhood school for the sake of
good community relations. Under these circumstances we often judge it a good thing to evince and show solidarity
with our children by putting ourselves in somewhat the same situation — e.g. share their diet or their exercise, or
become involved in the parent-teacher organization of the neighbourhood school — even if (as in the diet example)
there is no other good served by our doing so. It lessens somewhat the badness of having suffering imposed upon us
that he who imposed it suffers with us, especially, but not only, if we know about it and can thus derive encouragement
from it. (I write ‘not only* for the same reason as I have made a similar comment in connection with similar examples.
If we are glad when we learn about something, that can only be because we judge that what we learn about is
something good anyway, whether or not we learn about it.) The sharing of my suffering by my friend if that suffering is
for a good cause, and my friend suffers with me because he seeks to express solidarity and support, is a good which
compensates in part for the suffering. It makes, I suggest, the badness of the whole less than it would be if I alone
suffered. If God shares the pain and other suffering to which he subjects us for the sake of greater goods, that indeed
reduces the badness of the suffering.

the comment of the Exultet, the hymn of the Catholic Easter Eve liturgy, on the sin of Adam: e O happy fault which merited a redeemer so great an

This is

id of such a kind.’

Conclusion to Part III

In Part II I listed many good states of affairs which God in his goodness would seek to provide in a world which he
created. In Part III I have shown that so many of these good states could not be provided by God without his
providing certain bad states or the possibility of certain bad states which he would not prevent, at the same time.

The evils, moral and natural, which I have considered in Part III are, as far as I can see, all the ones there are, and of
each one of them it is the case that by allowing it to occur God makes possible a good which he could not make
possible without allowing it (or an equally bad state) to occur. Every moral evil in the world is such that God allowing it
to occur makes possible (given the assumption that humans have free will) the great good of a particular choice
between good and bad. Every bad desire facilitates such a choice. Every false belief makes possible the great good of
investigation, especially cooperative investigation, and the great good of some of us helping others towards the truth.
Every pain makes possible a courageous response (in all except animals caused to respond badly, and humans who do
not yet realize what is the good response), and normally the goods of compassion and sympathetic action. And those
animal pains to which animals are caused to respond badly, and those human pains to which humans respond with
self-pity in ignorance of what a good thing a courageous response would be, still provide many opportunities and
much knowledge for others. We can respond to the self-pitying humans by helping them to do better; their failure is
our opportunity. And all animal pain gives knowledge and opportunity for compassion to animals and humans if they
know of it. If the fawn caught by the fire in the thicket does not suffer, other deer will not so readily have the
opportunity of intentionally avoiding fire, he will not provide knowledge for other animals of how to avoid such
tragedies, other deer and humans centuries later will not be able to show compassion for his suffering, etc. (And if God
gave to angels a free will which allowed them to influence how the world was organized in respect to animal pleasure
and pain, then the possibility of animals suffering,

224

THE NECESSARY EVILS

unprevented by God, was a necessary condition of their choices making a difference.)

Why not let fawns suffer only if they are seen to suffer and we know about their suffering, can show compassion for
them, and can rescue them? But then from our observations of fawns suffering, either humans and deer draw the
correct conclusion — that fawns never suffer unless they can be seen to do so and can be rescued; or they draw the
wrong conclusion — that fawns do suffer when others do not know about it. In the former case we would then rightly
conclude that we need not worry about the possibility that some fawn will suffer for long in a fire which no one has
taken any trouble to prevent or avoid. And then a whole range of very serious choices will be closed to humans, and a
whole range of protective actions will be closed to deer (steering fawns away from possible fires from which those
caught cannot escape etc.). In the latter case God would violate the Principle of Honesty, and let deer and humans
show great compassion for fawns who never suffered; and take steps to prevent suffering which would never
otherwise have occurred. And that would involve God practising a massive and uncorrectable deception on his
creatures. I conclude that the suffering of fawns not readily rescuable is necessary for the great good of humans and
deer helping fawns, without God violating the Principle of Honesty. And we can have compassion too for these
anonymous but not completely unknown sufferers. As for the suffering of which we do not know at all: if we really do
not know of it (not merely do not know exactly when and where particular suffering occurred, but do not know that
there was any such thing as suffering of a certain kind), then we have no reason to suppose that it occurred and no
reason to take it into account in our theodicy. None of these goods could be provided without the cited bad state
unless God were to engage in massive deception: unless he let us believe that others are suffering when really they are
not, in order to give us the opportunity to show sympathy, and so on — a very bad state indeed.

Each bad state or possible bad state removed takes away one actual good. Each small addition to the number of actual
or possible bad states makes a small addition to the number of actual or possible good states. A reader might suggest
that no good would be served by his suffering a headache today, but he would be mistaken: it serves as a deterrent — to
studying too much, at the expense

CONCLUSION TO PART III

225

of doing other things; as an opportunity — for patience, not grumbling, etc.; as a source of information for the future,
to guard him against too much study in future, to provide information for his doctor to develop a theory about the
cause of the headaches more securely based than would be a theory with one less datum to support it. And so on. Of
course removal of one bad state or the possibility of one bad state will not remove much good, any more than the
removal of one grain of sand will make much difference to the fact that you still have a heap of sand. But the removal
of one grain of sand will make a bit of difference, and so will the removal of one bad state.

The crucial issues are whether all of the good states made possible by the bad states actually occur (in particular, do we
have free will so that our choices between good and bad are serious free choices?), whether a God has the right to
allow humans and animals to suffer in these ways for the sake of a good that that makes possible, and whether the
good is a great enough good for that purpose. I have argued that it is obvious that many of the good states (e.g.
compassion) occur; and that Christian revelation gives grounds for supposing that the other good states occur and it is
perfectly compatible with all else we know about the world to suppose that they do. To the issues of God's rights and
whether the goods are good enough, I turn in the final part.

This page intentionally left blank

IV Completing the Theodicy

This page intentionally left blank

12 God's Right

0 man, who art thou that repliest against God? Shall the thing formed say to him that formed it, why didst thou make
me thus?

(Romans 9: 20)

1 have argued in Part III with respect to all the evils, moral and natural, which occur on Earth, that if God allows them
to occur, thereby — on the assumption that certain theological doctrines, and especially the doctrine of human free will
and doctrine of life after death, are true — he makes possible various good states described in Part II, which would not
otherwise be possible and does all else he can to ensure their occurrence. It remains to be shown, first, that it is morally
permissible for God to bring about these bad states for the sake of good states which they make possible, i.e. that he
has the right to do so; and, secondly, that the expected value of allowing the bad states to occur is positive, i.e. roughly
that the goods which they make possible are at least a tiny bit better than the bad states necessary for them are bad. I
shall argue the former in this chapter, and the latter in the final chapter.

Talk of a ‘right’ can raise hackles. It suggests agents not concerned to be loving and generous, but insisting on their
rights. I must emphasize that I am using ‘a right’ in the sense that someone has a right to do an action, if and only if it
is morally permissible for them to do it, that is they do no wrong, i.e. do no one else a wrong by doing it. God has a
right to do something if and only if he does no wrong to anyone else by doing it.

230

COMPLETING THE THEODICY

The Rights and Duties of Carers

God is our creator. He gives us life and so many good things; and we are utterly dependent on him from moment to
moment, for he keeps operative from moment to moment the laws of nature which keep us in existence and provide
for us all those good things. We are God's creatures. God's relation to us is thus like that of human parents to their
children, not mere biological parents but parents who give their children nourishment and education from their earliest
years. As (to a limited extent, but more so than any other human) the source of being of the child, parents have certain
duties towards the child. To have brought a rational being into the world gives one a considerable duty, to give her a
good life. In so far as the duty is fulfilled, the parent has certain rights with respect to the child — a right to determine
how she shall flourish (e.g. where she will live and be educated); and also a right to certain lesser benefits from the child
in return (e.g. limited obedience while she is young but no longer an infant). Since God is so much more the source of
our being than our human parents (who can only give us what they give us because God keeps them in existence and
keeps operative the laws of nature which enable them to benefit us), he must, by analogy, have far greater duties and
rights than they do.

Such duties of caring for others may be possessed by others than a creator or semi-creator. Grown children have a
duty to care for aged parents (in consequence of the parents' previous care for them); and from that duty arise certain
rights; in particular, rights to determine how the parent is benefited (e.g. that the parent will have to move house nearer
to the son or daughter, in order that the latter may be able to care without too much difficulty). Most of us are
dependent to some lesser extent at various times on other persons and authorities, such as teachers and the state, the
latter having certain duties with respect to us, giving rise to certain rights over us. The greater the duty to care, the
greater (if the duty is fulfilled) the consequent rights.

The English language has a word for those who are the object of such rights and duties: dependants. Young children
are dependants of their parents, the incompetent aged are dependants of their middle-aged children; and all of us are
God's dependants. But English seems to have no general word for the ones on whom

GOD'S RIGHT

231

dependants depend. I shall use the word ‘carer’ in this defined sense (not quite the same as its normal sense) for those
who have a duty of care for others and rights deriving from their duty. Carers may often entrust decisions about
dependants to agents whom they regard as having the proper expertise to make them, e.g. doctors.

The principles which should guide human carers and their agents in making decisions about the future of babies and
young children, the senile and the mentally incompetent when they are in no position to do this for themselves have
been the subject of a considerable literature in recent years. There is a full philosophically orientated survey of the
issues and viewpoints in Allen Buchanan and Don Brock's book Deciding for Others . 166 The principles which begin to
emerge here provide useful analogies for the principles which should guide God in his relations towards his creatures.

Buchanan and Brock acknowledge that in general of course people have a right to determine their own
future — subject, I would add, to positive obligations to others or negative obligations not to harm others in various
ways. But to the extent to which they are incompetent, the good of self-determination has to be weighed against other
goods. To be competent, claim Buchanan and Brock, an individual needs to understand the alternatives and their
consequences and to have a minimally consistent and stable set of values. Patients, for example, being presented with a
choice between no treatment for some disease and one or two alternative treatments, need to understand the cost of
the treatment, the possible dangers, the likely outcomes, and to have a view about which outcome they want and how
much they want it — and a view which they do not change hourly. Buchanan and Brock also hint, though do not say
explicidy, that patients need to have values which are not obviously erroneous. They write of ‘mental illness’ ‘distorting’
a person's ‘aims and values ’. 167 But it is an ethical judgement that illness has distorted, as opposed to merely changed, a
person's values, and implies that incompetence may involve moral blindness. And that view is, I suggest, the correct
one. We are incompetent to some significant extent if we fail to understand the moral significance of alternative
actions. Autistic children who have no sense of right and wrong are to that extent incompetent.

166 (Cambridge University Press, 1989).

167 Ibid. 32.

232

COMPLETING THE THEODICY

Buchanan and Brock acknowledge that competence is a matter of degree — individuals may be more or less
competent — and varies with the kind of task for which the competence is being assessed. Competence is also, they
acknowledge, relative to importance of choice: if a choice is reversible (e.g. a choice to attend a certain school, or take a
course of medication) before its consequences are fully realized, a less high level of competence in assessing
consequences seems to be required. But if the choice is between one operation which may prove fatal and no
operation, then a higher level of competence seems appropriate. Given his less than full competence, Buchanan and
Brock hold, the good of the dependant's self-determination (to the extent to which he is competent) has to be weighed
against his well-being, as determined in other ways; he may be unable to choose at all or may need to be protected
from the harmful consequences of his choice. Carers or their agents will need to make choices on behalf of
dependants.

There are three criteria which have been proposed for determining these choices of carers. If the dependant was
formerly competent, then the first factor to be taken into account is any choice he made then, namely an ‘advance
directive’ that, should he become incompetent, so-and-so should be done. Secondly, carers should try to form a
judgement about how the dependant would have chosen if he had been competent. This ‘substituted judgement’
criterion raises problems about how the counterfactual (‘if he had been competent’) is to be interpreted. But the basic
idea seems to be that you suppose the dependant to be the same in his basic moral views (i.e. his views about the
overall goodness of states of affairs) and the extent to which he acts on those views, but to be properly informed about
the natural (i.e. non-moral) facts of the situation (what are his life prospects under different treatments etc.) and to
argue rationally (i.e. in accordance with objectively correct deductive or inductive procedures) from those premisses.
But even this clarification allows for spelling out in different ways, e.g. ‘the same’ may be read (for one formerly
competent) as ‘the same as he was (when competent)’ or ‘the same as he would have been now (if competent now)’.

Buchanan and Brock point out that the (American) law puts a restriction on the application of this criterion . 168 Carers
may not

Deciding for Others , 117-18.

GOD'S RIGHT

233

choose on behalf of an incompetent dependant to be so generous as to serve the interests of others at the expense of
the incompetent's own ‘basic interest’, i.e. to be so generous as to deprive the incompetent of basic necessities, even if
(they judge that) the incompetent would have been thus generous, if he had been competent. The reason Buchanan
and Brock give for this restriction is an epistemic one: the difficulty of judging how the dependant would have acted,
and so the risk of abuse by carers and their agents, unless the law incorporates an absolute prohibition. This reason for
the law incorporating such a restriction would not of course provide a reason for an omniscient and perfectly good
carer not to apply the ‘substituted judgement’ criterion in all its fullness. But I suggest that there is a further reason for
the restriction: unselfish decisions (i.e. ones which lessen an individual's own well-being from what it would otherwise
be) are not ones which others are entitled to take on behalf of someone else; they can only be made by him.

The third criterion which Buchanan and Brock discuss for reaching decisions about the future of dependants is that of
what is objectively in the individual's best interest, i.e. promotes best what is objectively the individual's own well-being.
(Of course there may be disagreement to be resolved at that stage about what does promote the individual's well-being.
And much of this book is concerned with the issue of what our well-being does consist in.) There is a certain not too
unnatural interpretation of the second criterion on which it collapses into the third. If we consider how the individual
would have chosen if he was properly informed about the moral as well as the natural facts of the situation, argued
rationally, and behaved rationally (i.e. was guided by his true moral views, to the extent to which doing so did not lessen
his own well-being), and was concerned only for his own well-being, then the second criterion collapses into the third.

I need to stress that the duty to benefit, to promote what is objectively in the individual's best interest, is a duty to
benefit on balance, overall. Parents have the right to cause or allow their children to suffer somewhat for the good of
those children, or of others. I do not have the right to cause some stranger, Joe Bloggs, to suffer for his own good or
that of Bill Snoggs, but I do have some right of this kind in respect of my own children. I may cause the younger son to
suffer somewhat for his own good and that of his brother. The right arises, I suggest, from the parent being the source
of much good for

234

COMPLETING THE THEODICY

the child which entitles him to take some of it (or its equivalent) back if necessary (e.g. in the form of the life having
bad aspects). If the child could understand, he would understand that the parent gives life, nourishment, and
education, subject to possible retraction of some of the gift (as well as an obligation to perform some services in
return). Parents do, for example, have a (limited) right to send their children to a neighbourhood school, for the sake
of cementing relations in the community, even though the children will be somewhat unhappy at that school (by
contrast with the private school to which they would otherwise go), in view of the many other good things which
parents provide for children. Often such rights (of carers over dependants) arise from duties of the carer towards other
dependants. I have the right to force an elder child to help a younger child cope with some difficulty when I cannot
help her myself directly, in view of my duty to help the younger child . 169 Among the rights of the carer, at any rate of
children and subjects of the state, is the (limited) right to obedience. Parents are entitled to tell the child to help in
various ways about the house. And a right to obedience gives rise to a right to punish disobedience: clearly a limited
right to take back compensation and more in whatever form the authority requires, a right which, we have seen, it is
often good should be exercised. The frequent exercise of that right could in theory make the overall package of benefit
provided by the carer a negative one. A severe state could take from a malefactor more than it gives him. But of course
that will only arise from the malefactor's choice.

The right of a carer who provides a life overall good for the dependant to cause some harm to the dependant is, I
stress, a very limited one. Clearly someone who rescues a child from poverty and starvation and gives him on the
whole a good life does not have the right in return to abuse him sexually from time to time. The obvious reason why
there is no such right is that any human has certain absolute rights, such as the right to choose how to use their own
sexual organs, by which the carer's rights are limited. While the obvious reason seems to me obviously correct, note
two important qualifications. The first is that while A may have a right to x, and so everyone including B has a duty not
to deprive A of x, it does not

ests can conflict with doing what maximises

s' obligations towards their other children, as well
ke precedence over if {Deciding for Others , 236).

l legitimate self-intere

the child's well-being, and sc

GOD'S RIGHT

235

follow that everyone else has a duty to interfere to stop B depriving A of x. Good though it may often undoubtedly be
that we interfere in distant lands to stop abuses, it does not follow that we always have a duty to do so. And for many
limited abuses of human rights in one country, people in other countries have considered not merely that they have no
duty, but that they have no right, to interfere. While some parental abusing of children clearly does impose a duty on
neighbours to interfere, more limited abusing does not impose a duty on people in distant lands to interfere. Exacdy
where the line is to be drawn on which abuse gives whom a duty to interfere is, as are all line-drawing matters, a tricky
matter, but we can — most of us — agree on some clear cases each side of the line. The second qualification is that the
carer's rights to cause harm or to allow others to cause harm to the dependant are greater if doing so is the means of a
greater good to (the dependant herself or) someone else. So often why people have no duty to interfere to prevent
limited abuses (e.g. to children) is because by not interfering they will in the end make the parents and their children
happier.

God's Decisions About the Unborn

What does our analysis of the duties and rights of human carers suggest about the criteria which ought to guide God in
his surrogate decision-making?

While the dependence of human children on their parents decreases with the passage of time, human dependence on
God remains total. But human competence to make decisions does increase with time. God has to make decisions on
behalf of humans (and animals) over a whole range of stages of their development, but I analyse only two decision
situations: the situation before there are any humans (and animals), and the situation when a given human on whose
behalf a decision may need to be made is as fully competent as ever a human becomes. I shall suggest answers for
these situations; clearly intermediate situations will call for intermediate answers. Let us begin with the all-important
situation, the first one. God has to decide to bodies of which kinds governed by which laws of nature to join souls.
Shall he make bodies which will give to their owners only thrills of pleasure, or ones which will also

236

COMPLETING THE THEODICY

give pains, and under which circumstances? This decision about our future I have pictured as being taken early in time,
but clearly one can think of it as being taken at each moment before each human (or animal) comes into existence, or,
alternatively, tunelessly. Since those on whose behalf this decision is taken do not yet exist, they are totally incompetent.
God has a choice of what sort of people to make and in what kinds of situation to put them. There is no scope here
for the ‘advanced directive’ criterion, nor for the ‘suspended judgement’ criterion, except in the sense in which it
collapses into the objective criterion. For there are as yet no persons with moral outlooks and patterns of behaviour;
there is therefore no meaning in talk about how (in the more natural sense) such persons would have chosen if they
had been able to choose. Only the objective criterion has application. God must choose to give each of us a life which
is objectively in our best interest. There is not, I suspect, such a thing as a maximally good life which God could give to
anyone: whatever life you describe, there could be a better. But it must be overall a good life . 170

God who is so much more the source of our being than are our parents has even greater duties than they do in respect
of us, for he has to choose the laws of nature under which we live — not merely on what affordable food we shall be
fed or by which accessible education we shall be educated. Consequently, his rights to choose how we are to be
benefited cover so much wider an area. But it must remain the case that God must not cause harm to us which is
uncompensated by benefit to us. Just what that amounts to in detail is, like all moral issues involving quantities of good
and harm, difficult to state in terms of any helpful formula but easier to illustrate by example. But the crucial point is
that God must not over time

Two recent theistic philosophers of religion who have made this point are Eleonore Stump and William Alston. ‘Undeserved suffering which is uncompensated seems clearly
unjust; but so does suffering compensated only by benefits to someone other than the sufferer’ (Eleonore Stump, ‘Providence and the Problem of Evil’, in T. P. Flint (ed.),
Christian Philosophy (University of Notre Dame Press, 1990), 66) . ‘Any plan that God would implement will include provision for each of us having a life that is, on balance a
good thing’ (William Alston, ‘The Inductive Argument from Evil and the Human Cognitive Condition’, repr. in D. Howard-Snyder (ed.), The Evidential Argument from Evil
(Indiana University Press, 1996), 111) . But his sentence continues, ‘and one in which the person reaches the point of being able to see that his life as a whole is a good for
him’ (my italics). I do not endorse this further clause. One reason why I do not is that it seems to rule out God giving life to animals who never do come to see that their life

GOD'S RIGHT

237

take back as much as he has given. He must remain on balance a benefactor. The overall goodness of the
package — the good states outweighing the bad — may arise from the presence in it of any or all of the good states listed
in Part II. The package is likely to consist not merely of many ordinary desires and their fulfilment, but also of
opportunities to make significant choices of good and bad, and above all of ways of being of use.

Note, however, that the qualification which I endorsed that carers cannot take on behalf of a dependant unselfish
decisions, in the sense of ones which lessen a dependant's well-being from what it would otherwise be, has no
application where the carer is God considering what sort of creatures to create. For ‘what it would otherwise be’ means
what it would have been in virtue of acts of other agents (animate or inanimate). But before God creates us with a
certain kind of life, there are no other agents who determine our well-being. Note also that it follows from my earlier
arguments that God's right to allow some other free agent to cause harm to us is likely to be greater than his right to
cause that harm direcdy. (He can allow wicked men to do things to us that mere inanimate processes will not cause.)
And it also follows that this right is greater if our harm is the necessary means to some good to the sufferer or others.
It will inevitably be the case that God will only cause harm for the sake of good, as I argued in Chapter 1 .

A parent does not always know for certain the effect of his actions on the child, e.g. what will be the effect of
entrusting the child to a normally reliable childminder, or even what it is likely will be those effects, through not
knowing some aspect of the childminder's character. But a parent has the subjective right to entrust the child if he
justifiably believes that the childminder will look after the child well. Given that, his subjective goodness will not be
affected by the fact that the childminder does not in fact look after the child well. But, we saw in Chapter 1, the perfect
goodness of an omniscient being requires him doing no objectively bad actions, and so only acts which he has an
objective right to do. But the objectively bad or wrong action whose badness derives from its consequences derives this
not from its actual effects but from the probable consequences detectable by a being who knows everything about the
state of the world before the action is done. Hence, although doing no objectively bad action, on the assumption that
God's knowledge of the future is limited by the libertarian free will which he gives to

238

COMPLETING THE THEODICY

humans, even God cannot know in advance for certain the actual amount of harm one individual will suffer at the
hands of another in a given situation. But since God is also omnipotent, he is able to ensure that no one remains in
such a situation for longer than he (God) chooses; and is able, if he so chooses, to compensate the sufferer in other
ways as and when he chooses. So God can ensure that everyone has more good than bad over the period of their
existence. There must also be, I suggest, other limits to the amount of bad that God has the right to allow one creature
to endure, not through their own choice: limits of intensity and length of bad state, including suffering.

If there is a God, he has ensured that there is a limit to the amount any human (or animal) can suffer on Earth (or
endure by way of other bad states), not through their own choice. The primary limit is provided by the safety barrier of
death. God only allows humans to suffer at each other's hands or by natural processes for periods of up to eighty years
or thereabouts. If the bad, in particular the suffering, endured by any individual during that period outweighs the good,
God does have the power to compensate that individual in an afterlife. The primary point of a good afterlife in the
Christian tradition is, as we have seen (in Chapters 6 and 1 1), that people shall enjoy the Vision of God in Heaven. But,
as we have noted, the Christian tradition has allowed for other kinds of good afterlife, namely Limbo. And the parable
of Dives and Lazarus represents the good situation of Lazarus in the afterlife not as reward, or where he naturally fits
because of his goodness, but as compensation for a bad life on Earth. 171

Still, plausibly, there are limits to the extent to which God ought to allow us to be harmed on Earth even if there is
eventual compensation in the world to come. Analogy with human parents would suggest that it would be wrong to
take too much from the child before giving much to it. And the New Testament parallels and teaching about life after
death, other than the story of Dives and Lazarus, are concerned more with its role as reward and punishment, or with
people living the lives for which their character fits them, than with its role as providing compensation. We shall come
in the last chapter to these delicate issues of just how much bad is

171 Luke 16: 19—31. See v. 25: ‘Abraham said [to Dives], Son, remember that thou in thy lifetime receivedst thy good things, and Lazarus in like manner evil things: 1
he is comforted, and thou art in anguish.’

GOD'S RIGHT

239

too much, given the overall good for the individual and others which results. Meanwhile, however, we must bear in
mind, with regard to suffering, that, because suffering is often a benefit for the sufferer in the opportunities for heroic
action and soul formation which it gives him, and because suffering is often the means by which the sufferer is of use
to others, the net negative weight of the suffering is not nearly as great as it would otherwise be. Often it is on balance a
benefit for the sufferer, as the writers of the Philo kalia among others have seen so well (see Chapter 9). But even if the
suffering is on balance bad for the sufferer, nevertheless our creator, if he has given us many other good things, has the
right to use us to a limited extent for the sake of some good to others. Kant was surely correct to emphasize that one
must treat individuals as moral ends in themselves and not use them for the good of others. But the latter phrase must
be interpreted as ‘on balance’. It is, I suggest, permissible to use someone for the good of others if on balance you are
their benefactor, and if they were in no position to make the choice for themselves.

God's Decisions About Adult Humans

We have seen that to the extent to which a dependant is not competent to make decisions for herself, the carer must
make decisions on her behalf; and the principle by which the carer must be guided is simply that of benefit: he must
ensure an overall balance of good (subject to any further qualification on the amount of bad involved in the overall
package).

But if dependants begin to become competent to make decisions for themselves (as animals never do), dependants
acquire an increasing right to determine which gifts they receive and which gifts (although objectively beneficial) they
do not receive. The rights of the carer become curtailed. The dependence of human children on their parents,
especially in our times their financial dependence, continues to a very limited extent long after children acquire
significant competence to make decisions for themselves. Although continuing dependence gives parents continuing
rights (and duties), clearly those rights (and duties) become less as children become more competent to make their own
decisions. To have

240

COMPLETING THE THEODICY

brought into the world a free moral agent, such as a progressively more competent human being, severely limits the
decisions one can take for him.

So now consider God's choice in respect of the well-being of adult humans such as ourselves as competent as humans
typically are. We remain as totally dependent on God as ever we were for our existence and the conditions thereof. In
virtue of the fact that we are currendy making (relatively) competent choices, the ‘advanced directive’ and ‘suspended
judgement’ criteria again have no application; only the objective criterion ought to guide God's surrogate decision-
making. But my arguments suggest that the right of carers to make decisions on behalf of dependants gets very much
less when, even though still dependent, the dependants become competent. Ought we not to be able to decide whether
we are to live in a web of interrelations in which our life is sometimes of use to others? Even if on balance our life is
good, if it is not what we choose, we have some right to reject it: a limited right in view of our continued status as
dependants (and our lack of perfect competence). That is surely so, and God must respect that right. And I hope my
readers will forgive my alluding again to the grim truth that we do all, when we become adult, unless prevented by
other humans, have the natural possibility of not continuing to live the sort of life which God makes available to us.
We can all commit suicide. If we do not like our hand of cards, we can hand it back. Babies, animals, and young
children may not be able to commit suicide; but of course they are not competent to decide whether to do so.

True, as I wrote earlier, we are all under a moral obligation not to commit suicide. We may be under such an obligation
to parents and children and others who love us not to commit suicide. Such obligations exist either as a result of our
own choice (we have chosen to have children) or as a result of factors outside our control (we have not chosen our
parents). God cannot remove these obligations except by killing off those who love us or by making them love us no
longer. Yet it seems implausible to suppose that a perfectly good God would do either of these things in order to
remove our obligation not to commit suicide. And anyway those who are obliged not to commit suicide because they
are loved are fortunate — to be loved. So many of those who decide to commit suicide decide to do so because no one
loves them. And, secondly, we are under an

GOD'S RIGHT

241

obligation to God himself not to commit suicide. On the assumption that God ensures that the gift of life is on balance
objectively a good gift, no human ought to throw it back in the face of the giver. God is under no obligation to remove
that obligation, for logical reasons; for B cannot be under an obligation to A if A is obliged to let B off the obligation if
B requests; for B's obligation then amounts to nothing. Nor would it be good for God to let us off our obligation not
to perform what would be objectively a bad act. I conclude that a perfectly good God would not remove any of our
moral obligations not to commit suicide. But we do have the power, if we so choose, to commit suicide (wrong though
it would be), unless, that is, others prevent us.

Ought not God to have kept from humans the power to stop other humans committing suicide? But that would have
kept from humans the power to frustrate other humans (for a while) in doing themselves ultimate harm. Having this
power is a great good for the former humans. Since what the would-be suicide is about to do is a great wrong (to his
creator and many others), and since the lengthening of life can be at most for a few years, and the would-be suicide
might well change his mind about his decision later, I cannot see that God wrongs the would-be suicide if he does not
stop others interfering in this way. For we humans recognize that we have the right to allow some humans to stop
others from committing suicide for limited periods. And, as always, God's rights are greater in quantitative terms than
ours, and may reasonably be supposed to extend to allowing some to prevent others from committing suicide for a
limited number of years, i.e. until death. We note that God himself allows us (if others do not prevent it) to commit
suicide very much sooner.

The conclusion of this discussion of God's right to allow humans (and animals) to suffer is that God does have this
right so long as the package of life is overall a good one for each of us. Bad aspects have to be compensated by good
aspects. But the results of Part II, that being of use is a great good, suggest that our normal estimates of the worth of
the package which do not take that into account are likely to be wildly in error; and that most lives which seem (unless
we take that into account) to be bad on balance are not really so. And note one final highly important point about being
of use. If the world was created by God with all the providential purposes which this book has been setting out when
we are of use in forwarding

242

COMPLETING THE THEODICY

these we are of use to God himself. Puny humans and even punier animals are given a role to play in his plans by the
all-good source of all things. Being of use is indeed a privilege. Yet if there are any lives which nevertheless are on
balance bad, God would be under an obligation to provide life after death for the individuals concerned in which they
could be compensated for the bad states of this life, so that in this life and the next their lives overall would be good.

One of the most frequent ways in which people express their protest against the world's ills is by saying that what
happens to them or to someone else is not ‘fair’. Lying behind this remark is often the feeling that God would only be
justified in causing some ill as a punishment for wrongdoing. This feeling springs from a very narrow vision of which
states are good and which are bad. Life, even a life which includes much bad, is in itself a great good; and the bad
provides the opportunity for choice and being of use. The good things and the bad things of life are indeed distributed
unevenly (see p. 149); to each is given a different package of various things, good and bad, to be used in different ways.
Thus God treats us as individuals, each with her own vocation. This would only be ‘unfair’ if the bad states were too
bad or not ultimately compensated.

13 Weighing Good Against Bad

Remember the words of the Lord Jesus, how he himself said, it is more blessed to give than to receive.

(Acts 20: 35)

I reckon that the sufferings of this present time are not worthy to be compared with the glory which shall be
revealed to usward.

(Romans 8: 18)

I argued in Chapter 1 that to produce a theodicy for a certain bad state E one would have to show that
(a) God has the right to allow E to occur.

(B) Allowing E (or a state as bad or worse) to occur is the only morally permissible way in which God can bring
about a logically necessary condition of a good G.

(d) The expected value of allowing E, given (c), is positive.

I claimed that in order justifiably to believe that there is a God, despite the existence of bad states, one needs to have a
theodicy for the various kinds of bad states which there are. Or rather that one needs this in the absence of very strong
evidence for the existence of God, or a record of discovering with respect to a succession of kinds of bad states that
one could in the end always find a theodicy for each one considered. Because this book is an examination of the
philosophical issues raised by Christian theism, as opposed to theism generally, I have added to the central Christian
claim that there is a God various other Christian doctrines (such as the occurrence of and conditions of life after
death), and I have been examining how the whole package faces the difficulty that God allows so many awful things to
happen.

244

COMPLETING THE THEODICY

I argued at the end of Part III that condition (B) is formally satisfied for each of the world's bad states. Each one is
such that allowing it or an equally bad state to occur is the only morally permissible way in which God can make
possible the occurrence of some one of the good states described in Part II. Given our theological assumptions, the
good states which the bad states make possible are realized. Or, to be more precise, where the good states are such
that it is logically possible that God bring them about, they are realized; where this is not logically possible (e.g. where
the good is some agent freely doing some action), then God does all else needed (besides allowing the bad state) which
is logically possible for him to do to facilitate the occurrence of the good state. If the good state is my freely choosing
the good rather than the bad, God does his part in giving me free will. So condition (c) is also formally satisfied, for
each bad state, as the reader can check by running through the bad states listed in Part III and the ways there described
in which they make possible the good.

It is often the case, especially given our theological assumptions, that bad states make possible many different good
states. Some one bad state may be needed for various different goods, each together with some other bad states
different in each case. It might be that the goods are none of them individually good enough to justify God bringing
about a particular bad state, but all the good states together for which all the various bad states are needed are good
enough to justify God allowing the bad states to occur. There is no general explanation why God allows the bad, but a
range of explanations, some one or more of which have application to each bad state. So the Gs in the above formula
may include many good states; and (r) may involve allowing other bad states.

With respect to condition (a), we saw in the last chapter that God has the right to allow a bad state to occur to an
individual if the life which he gives to that individual is on balance a good one — and the bad periods are not too long
or too bad . 172 Given the assumption of

In Chisholm's terminology God has the right to allow a bad state to occur to an individual only if that bad state (together with the other bad states in the life of that
individual) is ‘outweighed’ (or more than ‘balanced off’) by good states in the life of that individual. In ‘outweighing’, there are separate good and bad states, such that it is
logically possible that the one occur without the other. But in virtue of his perfect goodness God will only allow a bad state to occur if (probably) allowing the bad state
forms part of a good whole, such that it is not logically possible that God could bring about the good without allowing the bad. In such a case, in Chisholm's terminology,
allowing the bad is part of a whole in which it is ‘defeated’ by God bringing about the good. But in this case the bad may occur to one individual, and the good to another.
Very loosely, every bad must be defeated by some good; but in the life of one individual the bad needs only to be outweighed by some good. See Roderick M. Chisholm,
‘The Defeat of Good and Evil’, repr. in M. M. and R. M. Adams, The Problem ofTivil (Oxford University Press, 1990) .

WEIGHING GOOD AGAINST BAD

245

life after death for humans, the compensation for bad periods may for some of them come after death. Whether the
bad human periods are too long or bad, and whether an animal's earthly life is on the whole good, depends then on
quantitative issues. And whether condition (d) is satisfied is entirely a quantitative matter. So the total quantitative
requirement is that the expected value of allowing any bad state to occur should be positive; and if it is negative for a
particular individual there must be enough compensation for that individual in this life, or the next life. So everything
turns on the balance, or, rather, given that humans may act contrary to the objective probabilities, the probable balance
of good over bad, not merely as a whole, but in the life of each individual.

This issue of whether the goods are great enough to justify the bad states which make them possible is the crux of the
problem of evil. I have found in discussion of these matters that my opponents are usually happy to grant me, when I
bring the suggestion to their attention, that the states which I describe as ‘goods’ cannot be had without the
corresponding bad states, and quite often happy to grant that the former states are indeed good states and even that a
world is not on balance worse for containing a few of these goods in the mildest of forms together with the necessary
bad states than it would otherwise be. But my opponents usually object to the scale; there are, they claim, too many, too
various, and too serious bad states to justify bringing about the goods which they make possible. There would, as I
wrote earlier, be something wrong with one morally if one did not have some initial sympathy with that objection. This
world's ills are many and tragic, and a conception of ‘goodness’ which did not lead one to suppose that a good being
would try to eliminate them would be a corrupt one. And the requirement that for each individual there be a probable
balance of good over bad in his life if God is to be justified in creating him looks initially hard to meet.

How are we to weigh the bad states against the good states which they make possible? Any philosophical consideration
of the relative merits of these good and bad states should perhaps begin with the

246

COMPLETING THE THEODICY

comment of Plato, through the mouth of Socrates, in the Gorgias, that it is worse to do injustice than to suffer it . 173
Spurred by that to my mind profoundly correct insight, we need to test it and similar insights by reflecting at length (as
well as on our actual life experiences) on innumerable very detailed thought experiments in which we have a choice of
different kinds of life to live, and thereby to draw out what sort of life we regard as on balance good for us; and then
also by reflecting at length on detailed thought experiments in which we to some extent play the role of God and have
a limited choice of worlds to create, and have to choose the best (or equal-best) of this limited set. These thought
experiments enable us to draw out our comparative judgements. Our judgements may be mistaken, but they are the
ones which must guide us until more experience and reflection leads us to judgements which we regard as better. To a
considerable extent much of what I have already written in this book has been by way of thought experiments
suggestive of our actual judgements about the great or little value of certain good and bad states. But I must now begin
to draw out some comparative results, by means of a few further thought experiments.

Let me begin by emphasizing the goodness of life itself. Suppose that, throughout your life, you have available a
machine by pressing a button on which you can become unconscious during the periods of pain, mental agony, and
even boredom. Pressing the button will make you unconscious for an hour or two, during which you behave as though
you were conscious, and after becoming conscious again you know what happened in the meantime. If when you
become conscious again, you do not like what you then find yourself experiencing, you can go on pressing the button
until you find your life more to your liking. Periods eliminated will not be replaced, and so pressing the button will
shorten your conscious life, for as much or as little as you choose. How many of us would press the button for long?
Not many, I suggest, would press the button very often. And that brings out that most of us value simply existing as
conscious beings, whatever (within limits) life throws at us. We Value’ it, in the sense that we recognize it as objectively
good for us.

To bring out this point further, I hope my readers will forgive me for mentioning again the grim fact of the possibility
of suicide, but

WEIGHING GOOD AGAINST BAD

247

now more positively Very few humans indeed commit suicide, although almost all of them could do so quite easily.
True, some of them do not do so because of obligations to others, especially I suspect often to their children. But
many others do not commit suicide because they want (i.e. desire) to go on living, even when life is unexciting or
painful. One reason why they so desire is that they think that — unwilling though they often are to admit it, when badly
depressed — the good outweighs the bad. They must therefore think that merely going on having experiences is a good
which outweighs so much bad. Life is a tremendous good in itself. The other reason why they desire to go on living is
that, even if they think that at present the bad outweighs the good, they live in hope of better times. Thereby they
express their belief that a life good as a whole over time would be worth having even if its present state is on balance
bad.

Throughout this book I have emphasized the great goodness of having free choice, and being of use to myself and
others. It would, I suggest, be very difficult to construct a satisfactory theodicy which did not rely on the doctrine of
human free will. Although some of the strands which I have deployed earlier in the book do not rely on it — e.g. some
animal suffering may be justified by the good of being of use, even if the use is not freely chosen — the worst states of
the world are cases of suffering (not just sensory pain, but pain including shame and humiliation) which humans
deliberately and maliciously inflict on other humans together with the evils of their inflicting them. Here the moral evil
is so great that we need (as well, often, as other good states), in order to justify the resulting bad, the good of those
humans (and of those who allow them to get into the state of having the power and inclination so to act) having a free
choice of deep significance.

But free will will not suffice for theodicy; and the other great good which I have stressed is the good of being of use.
All the ways in which the suffering of A is beneficial for B are also beneficial for A — because A is privileged to be of
use. The fawn caught in the burning thicket is privileged to be of use to other deer; and we have seen many times that
the fact that the sufferer does not realize the benefits he is conferring and the privilege of conferring them does not
affect the fact of the privilege. The fawn's life is of use; it is not wasted — he has enabled others to save themselves.

We must not overestimate the intensity of animal suffering, I argued earlier. But when we come to human suffering,
the following

248

COMPLETING THE THEODICY

kind of thought experiment may help us to see the value of the great privilege of being of use. Suppose that you exist in
another world before your birth in this one, and are given a choice as to the sort of life you are to lead. You are told
that you are to have only a short life, maybe of only a few minutes, although it will be an adult life in the sense that you
will have the richness of sensation and belief characteristic of adults. You have a choice as to the sort of life you will
have. You can have either a few minutes of very considerable pleasure of the kind produced by some drug such as
heroin, which you will experience by yourself and will have no effects at all in the world (e.g. no one else will know
about it); or you can have a few minutes of the considerable pain of childbirth, which will lead to the existence of a new
human being who would not otherwise exist. Because the latter human being will never exist unless you choose the
childbirth alternative, you are under no obligation to bring about his or her existence: no one is wronged if you make
the other choice, for there is no one to be wronged. (The celibate wrong no one by not procreating.) But you seek to
make the choice which will make your own life the best life for you to have led. How will you choose? The choice is, I
hope, obvious. You must choose the second alternative.

We can make various alterations in this example, which will all confirm this great value of being of use. The pain may
be a pain which causes not the existence of a human but that a human who would otherwise be a very happy zombie
becomes capable of moral choice, and desires knowledge. You do not wrong the zombie if you do not elevate his
status: he has, let us suppose, no other connection with you; and, we assume, is very happy. But you seek to benefit
him: to make him no longer just a recipient object, but a choosing, agonizing, knowledge-seeking agent. If you take
Mill seriously, ‘better to be Socrates dissatisfied, than a fool satisfied’, and you want the zombie to have a good life, you
will choose again the pain. It would of course be irrelevant to your choice whether its effects were near or distant in
time and space — if like Christ, or perhaps the angels, or Adam, you are making a choice whose effects are far away.

Now suppose you have a God-like choice between creating our world; or a world where all pain is 10 per cent of its
intensity in this world, people only suffer it for at most eight years (though they live as long as in our world), where evil
desires are only 10 per cent in frequency and intensity of those in our world, and people only have

WEIGHING GOOD AGAINST BAD

249

10 per cent of the choices and can only make 10 per cent of the differences which they can in this world. Otherwise the
world is programmed to move along in much the way our world does. There are obvious advantages in such a
world — who would deny that? But, as regards the kinds of good to which I have drawn attention in this book, it would
be a toy world. Things would matter in the kinds of respect which I describe, but they would not matter very much. It
would not matter very much if we made the wrong choices, because not very much would depend on them. If we did
not feed the starving, clothe the naked, use force to establish justice, and give love to our children, it would
matter — but it would not matter like it does in our world. For most of the starving would get food even if we did
nothing; and although some might die, they would not feel much pain. And the naked would not be much worse off
than we are on chilly days when we have forgotten to take a coat with us. People would behave in a fairly just way
anyway without the need for much force, and a few swings of a truncheon would intimidate the average wrongdoer.
And our children, though valuing our love if they got it, would not mind too much if they did not get it. Is it really so
obvious that God would be less than perfectly good if he gave us a world where things matter a lot more than that? Of
course not.

I must emphasize yet again that I am not saying that a world with a lot more choice and a lot more opportunity to be of
use, together with the bad states which would be needed for that, would be better than our world. There certainly does
come a point where additional bad states make things overall worse, and a point at which it would be quite wrong of a
creator to create a world with so much bad in it. But I am suggesting that thought experiments show that it is not clear
that we have reached that point with our world. Nor, I emphasize once again, does it follow from God having the right
to create such a world as ours that we have anything like the right that God has to cause someone to suffer for the sake
of some good for him or others: God's right to create, cause, and allow suffering arises from his being from moment
to moment the source of our being.

Some writers have claimed that certain sorts of good states, say of kind x, are not comparable in goodness with other
sorts of good state, say of kind j, in that, however many prolonged states of kind j you have, it will still be better to
have one state of kind x. Noah

250

COMPLETING THE THEODICY

Lemos, in an important article 174 on the comparison of kinds of goodness, cites Parfit, Brentano, and Ross as explicit
defenders of this view, and Aristotle and Mill as giving it some oblique support. Thus Ross writes: ‘with respect to
pleasure and virtue, it seems to me to be much more likely to be the truth that no amount of pleasure is equal to any
amount of virtue, that in fact virtue belongs to a higher order of value, beginning at a point higher on the scale of value
than pleasure ever reaches ’. 175 These writers hardly have a religious axe to grind.

I share with Lemos the opinion that Ross's particular claim is implausible: an enormous amount of pleasure would equal
a very small amount of virtue. But Lemos uses the general idea behind Ross's view to urge that the goodness of a life
of connected goods which may indeed be far better for including ‘pleasure, consciousness, and knowledge’ would, if it
was a woven-together and integrated finite life, be better than an infinite amount of pleasure. That seems to me
correct, and gives more plausibility to the suggestion that a finite life of painful utility is preferable to an infinite life of
low-grade kicks of pleasure, the former, say, gaining its unity from a series of hard experiences which give to others
opportunities for forming their souls through choosing between good and bad. Again if before birth I had the choice
between lives of these two kinds, though I doubt if (being the sort of person I am now) I would choose the former life,
I can understand it being the better life. But of course I cannot choose, and God has to make the choice for me. God
sometimes pays us the compliment of supposing that, if we had the choice, we would choose to be heroes.

Our culture has dulled our moral sensitivities to the goodness of free will and (especially) to the goodness of being of
use (and of much else — of honesty, fidelity, the sanctity of life, truth-telling, and promise-keeping), and overestimated
the goodness of mere pleasure and the badness of mere pain — and maybe not even earlier centuries have fully
appreciated the former moral ultimates. Yet even if an opponent allows the formal point that there is great value for
the subject in being of use and being helped, he may fail to see that that has the consequence for theodicy which I
commend, because of two characteristic human vices: short-term and short-distance

174 Noah M. Lemos, ‘Higher Goods and the Myth of Tithonus’, Journal of Philosophy, 90 (1993), 482—96 .

175 W. D. Ross, The Right and the Good (Clarendon Press, 1930), 150 .

WEIGHING GOOD AGAINST BAD

251

thinking. He tends to think of the worth of a sentient life as dependent on things that happen during that life and fairly
close in space to the life. But once you grant the formal point that things outside a life, e.g. its causes and effects, make
a great difference to the value of that life, it seems totally arbitrary to confine those things to ones near to the life in
space and time.

To take another particularly unpleasant example: the eighteenth-century slave-trade. The white slavers captured,
inhumanely transported, and sold Africans into American slavery only because over centuries they had been taught to
despise black people as uneducated savages, and had been taught that there was nothing wrong in enriching yourself at
the expense of the suffering of uneducated savages. Innumerable parents, preachers, and politicians had their share in
creating such a culture; and they deliberately chose to ignore the moral worth and human feelings of black people; for
such a culture marked a decline in human standards: it did not exist at the beginning of the Middle Ages. But God
allowing this to occur made possible innumerable opportunities for very large numbers of people to contribute or not
to contribute to the development of this culture; for slavers to choose to enslave or not; for plantation-owners to
choose to buy slaves or not and to treat them well or ill; for ordinary white people and politicians to campaign for its
abolition or not to bother, and to campaign for compensation for the victims or not to bother; and so on. There is also
the great good for those who themselves suffered as slaves that their lives were not useless, their vulnerability to
suffering made possible many free choices, and thereby so many steps towards the formation of good or bad character.
And for the victims there remain the possibilities of compensation and reward after death.

Note that in so many such horrible cases the free choices often became available to people too hard-hearted (as a result
of previous bad choices) to do normal acts of kindness (e.g. to Africans). A good God will be desperately anxious to
rescue the hard-hearted before they become incorrigibly hard-hearted, and to rescue them means to help them to make
choices which will put them on the road to sanctity. But (see Chapter 9), the choices for the hard-hearted who have
immunized themselves to moderate amounts of suffering in others can only be ones where the wrongness of the bad
choice is very evident and very great. But to repeat quickly — yet again before anyone misunderstands — only God our
creator had the right to

252 COMPLETING THE THEODICY

allow bad people to promote the slave-trade. Humans had the duty to fight very hard against it; and a good God would
very much want them to do that. Yet I cite this example, among so many others which could be cited, to illustrate how
the possibility of large-scale human suffering opens up innumerable opportunities for significant choices in earlier
centuries and distant places, and how its occurrence opens up innumerable opportunities for good or bad choices of
response to it in later centuries and distant places.

If we continue to reflect on many other thought experiments similar to those of this book — how we should choose
between other alternatives with longer lives or different lives (incarnation as a fawn or a suffering child 176 maybe)
against a background of many centuries of effect and cause and place in the web of human and animal society, we may
begin to look at things a litde more sub specie aeternitatis , and that, I suggest, will lead us to the kind of assessments of
relative value which I am commending. But of course our moral judgements are subject to limited error, especially
when the comparison of goods is involved. God may reveal to us moral truths, or at any rate some details of moral
truths (including the comparative worth of good states) which we are not able to discover for ourselves. If we have
evidence from the circumstances of the foundation of a church which proclaims a purported revelation that God has
indeed given us a genuine revelation, we will need to amend our moral judgements. But it will remain a major test of
whether a purported revelation is genuine that it fits with (although goes beyond) our secular knowledge — and that
includes the general oudines of our knowledge of what is good and what is bad. Given that the arguments and thought
experiments which I have given do suggest that free will and above all being of use are great goods, we can appeal to
the Christian revelation to confirm that they and especially the latter are even greater goods than natural reason alone
suggests.

The New Testament on the Value of Service

That service of others, humans and God, voluntary or involuntary — being of use to them — is a tremendous good for
the server

176 See Ivan's speech in F. Dostoevsky, The brothers Karamazov ,

book 5, ch. 4.

WEIGHING GOOD AGAINST BAD

253

seems to me something very close to the heart of the New Testament. I have already cited St Paul's report of the words
of Jesus: ‘it is more blessed to give than to receive’. But let us run through other passages in the New Testament which
convey the same message. I begin with the words of Jesus to his disciples on what their greatness is to consist in: ‘Ye
know that they which are accounted to rule over the Gentiles lord it over them; and their great ones exercise authority
over them. But it is not so among you: but whoever would become great among you shall be your minister: and
whosoever would be first among you shall be servant of all. For verily the Son of Man came not to be ministered unto,
but to minister, and to give his life a ransom for many .’ 177 The passage classically connects greatness with service, and it
is most plausibly seen as saying that greatness consists in service. And, analogously, while reminding his disciples that
he is ‘Master and Lord’ and that ‘a servant is not greater than his Lord’, Jesus washed his disciples' feet; suggesting,
though not explicitly stating, that service is part of greatness . 178 In several places the New Testament teaches that those
who suffer in consequence of their choice to confess the name of Christ are fortunate to be allowed to have such a
significant role in the proclamation of the Gospel. The Aposdes beaten for preaching the Gospel rejoiced ‘that they
were counted worthy to suffer for the name ’. 179 St Paul wrote to the Colossians that he ‘rejoiced’ in his sufferings for
their sake . 180 St Paul's reward for preaching the Gospel was to preach the Gospel — without being paid for it . 181
Conversely, the sentence (xpaic;) on those that did not believe ‘on the name of the only begotten son of God’, is,
according to Jesus as St John reports him, ‘that the light is come into the world, and men loved the darkness rather
than the light; for their works were evil ’. 182 Good and bad actions are their own reward and punishment — which is not
to deny that they may obtain other rewards and punishments as well. The Gospel teaches both a present reward for
doing good and a future reward; and inevitably the two get a bit mixed up in its message. A passage which tells us of
both kinds of reward is the Beatitudes, as St Matthew records them in the Sermon on the Mount . 183 The blessing for
‘the pure in heart’, for example, is said to be a reward: ‘they shall see God’, which may

Mark 10: 42-5.
John 13: 1-16

180 Col. 1: 24.

181 1 Cor. 9: 18.

John 3: 19.
Matt. 5: 1-12.

254

COMPLETING THE THEODICY

plausibly be thought to lie in the more distant future; whereas the blessing for ‘the peacemakers’ is the reward of being
‘called sons of God’, which is something which may plausibly be thought to be available sooner. (Emerson wrote,
more prosaically, that ‘the only reward of virtue is virtue ’; 184 but the word ‘only’ seems to me to put his remark out of
line with the New Testament.)

And there is a long Christian tradition that the suffering and death of Christ, God incarnate, constituted his glory. ‘He
reigned from the tree,’ wrote Venantius Fortunatus (early seventh century AD) in the well-known Passiontide hymn
‘The Royal Banners forward go’. God is glorified in his suffering and death on the cross quite independently of what
might follow for him or for us. To suffer for our sake (however that is spelled out in a doctrine of the Atonement) is
his glory. But of course after suffering he entered ‘into his glory 1185 of a different kind. And if the words of dereliction,
‘My God, my God, why hast thou forsaken me ?’, 186 are to be taken in any way seriously, Christ's moment of supreme
service was a moment at which he did not in his human mind realize that his suffering had the value it did. I suggest
that if we give to being of use by our suffering or our action, determined or free, the kind of value which the New
Testament suggests that it should have, and if we accept the other claims of Christian revelation (e.g. about free will
and life after death), the comparative condition is met. Each life, animal and human (given life after death for the
latter), is a good life — whether its goodness comes from pleasure or service; and every bad state facilitates a great
enough good — often in terms of pleasure and much else good for others.

The reader will be sympathetic to the theodicy if he or she thinks that it matters more — a lot more — that we should be
good people than that we should be happy people, though by far the best state of all is that we should be both; and that
while pleasure is a good, generosity is a greater good; and that while pain is in itself bad, being allowed to suffer to
allow others to exercise higher-order goods is a privilege which outweighs that bad. A loving God will give: like a
loving parent, he will make his younger children be useful; and he will allow his older children freely to choose whether
to be useful. And he will provide an abundance of opportunities for cooperative mutual benefit both in providing
tingles of first-order

Essay on ‘Friendship’.
Luke 24: 26.

Matt. 27: 46.

WEIGHING GOOD AGAINST BAD

255

pleasure as well as the effort of second-order generosity. The dark spots of bad give opportunities for particular kinds
of good.

The Primary Point of Life

It remains the case, however, that bad is bad, and there is a substantial price to pay for the goods of our world. God
would not be less than perfectly good if he created instead a world without pain and other suffering, and so without the
particular goods which they make possible. While permitting suffering for the sake of a greater good does have its
point, this is the sort of thing a parent or a king dedicated to the happiness of his subjects would do only temporarily
and for some subjects. It would be getting sacrifice out of proportion to let it rip uncontrolled, saying Tf it serves a
good, the more the better’. I have argued previously that God has the right to allow individuals to suffer only for a
limited duration. If that is so, I cannot see that he does not have the right to allow an unending stream of individuals
each to suffer for a limited duration. I suggest, however, that it is a good which God, as perfectly good, would seek to
ensure — that the total should be of limited duration. This exercise with animals and humans should have finite limits.
Christianity at its beginning firmly taught, and later Christian teaching (Catholic, Orthodox, and Protestant) has
continued to teach, that this whole world-order in which there is suffering on Earth is due to come to an end. This
world-order is a very dangerous and costly experiment which its author might be expected to bring to a halt one day.
Although the divine time-scale is longer than we readily grasp, the goodness of an end to the experiment remains. And
of course in the last thirty years physical cosmology has come to teach that probably the world will end in a finite time,
even if God does not intervene to bring it about earlier, through natural causes, either in the cold dark of a dying
universe or in the Big Crunch. That it will end does not of course serve by itself to justify the experiment in the first
place, but again it does mean that its bad states are limited ones.

Neither the New Testament nor subsequent Christian doctrine contain more than parts of a full-blown theodicy and
hints on how to construct one — hints which I hope I am taking into my own

256

COMPLETING THE THEODICY

construction. Subsequent Christian theology has usually given the central place in its theodicies to the ‘free-will
defence’, the virtual inevitability of wrongdoing and suffering if humans are to have the great good of free and
efficacious choice. I follow the tradition in this central respect. I think that some Christian tradition has
overemphasized the role of the Fall (see pp. 108-10), and under-emphasized the great good of being of service. And,
as I pointed out in Chapter 2, the point made most firmly of all at many places in the New Testament is the point just
made about the temporary nature of anything bad which is not the result of free choice. It has been allowed to occur
by God, the New Testament always emphasizes, for some good purpose (not always mentioned), temporarily. The
New Testament writers saw the healing miracles of Jesus and his disciples as a sign that the good purposes served by
natural evil were being fulfilled, and the natural evil was coming to an end. 187

And for its individual inhabitants too this world is temporary. Although what humans and animals do and suffer in it is
very important in itself, quite apart from its consequences, nevertheless, the Christian tradition also firmly teaches, the
primary point of (at any rate, human) life in this world is the next world. The primary point of the creation of the
human race is to give it the vision of God oudined in Chapter 6. Given the availability of life after death of this kind,
the bad states of frustrated desire for immortality and the vision of God, and the other bad states this world contains,
need only be temporary; and will be only temporary for those humans who choose to seek this vision on this Earth. A
major task for theodicy is, then, the task of justifying the vision not being given immediately, and my Chapter 11
sought to provide the theodicy for that — to show that it was good that the choice be affirmed over the course of a life,
possibly through our reaction to suffering. The alternative to Heaven, Hell, is primarily damnation (the supreme
penalty of damnum , the permanent ‘loss’ of the vision of God). I have argued earlier that it is good that free agents
should have the choice of finally rejecting the good, and so God and all that he stands for. But that of course is not
what God seeks for us. He seeks to take us to himself in Heaven, a marvellous world with a vast range of possible
deep goods, but one lacking a few goods which our world contains, including the good of being able to choose to
reject the good.

Luke 10: 17-18.

WEIGHING GOOD AGAINST BAD

257

The great goods of Heaven, as pictured in the Christian tradition, include that deep awareness of God which,
combined with even moderately good desires, logically rules out any of its inhabitants having the good of being able to
choose to reject the good. This deep awareness is an awareness of the ultimate source of things, the perfectly good,
loving, and beautiful God to whom the Blessed cannot but respond in worship. There are so many different ways of
worship open to them. Cooperative worship is depicted as the primary activity of the Blessed in the one book of the
New Testament which gives much attention to their fate. The Blessed share in the divine life of God; but they also
share in the divine work of God. The citation which I give from the Book of Revelation, at the beginning of my
Chapter 6, describes them as ‘reigning’ for ever and ever. There is no end to possible good states: there can be more
creatures; they can be given more powers, more knowledge, more choices. And the Blessed can help in the work of
love in forwarding these good states for themselves and others (including others still on this Earth). They could have a
choice between so many alternative good actions — between equal-best actions, or, perhaps for them as not for us on
Earth, 188 a choice between an infinite number of good actions, each one less good than some other action they could
do. The range of free choice open to the Blessed could be enormous compared with the range of choice open to us on
Earth.

A generous God might well choose to give some of us the choice of rejecting the good in a world like ours, before
giving to those who accept it a wonderful world in which that possibility no longer exists.

We saw in Ch. 5 n. 4 that a being who has power unlimited in some respect may have a choice of an infinite number of actions, each one less good than some other action.
For such a being there is no best or equal best action. God is in this situation, for his power is unlimited in all respects; we humans on Earth are not in this situation, for our
power is limited in all respects. But maybe God gives the Blessed unlimited power in some respect (e.g. to create as many atoms as they choose; or to paint as large a picture,
or sing as long a song as they choose). For the Eastern Orthodox tradition Heaven involves ‘deification’. By ‘deification’, as I noted earlier, the early Fathers simply meant in
some way coming to share the divine life. But for St Maximus the Confessor, whose championing of this notion led to its prevalence in the Orthodox tradition, deification
meant ‘becoming all that God is, except for an identity in essence’ (Maximus, Book of Antiquities , 41; PG 91: 1308). A natural way of interpreting this is to say that the
Blessed are — not by nature, but by God's grace — omnipotent, omniscient, and perfectly good. The Blessed would then have powers unlimited in all respects. But for them,
as for God, omnipotence would be constrained by perfect goodness, so that it would only be exercised in ways compatible with other agents doing what is good for them to

This page intentionally left blank

Additional Notes

1. [Note 1 to Chapter 1.] For a full spelling-out of what is involved in this understanding, see my The Coherence of
Theism, rev. edn. (Clarendon Press, 1993). All theologians of the Western church from the fourth century until the
fifteenth century AD, and most of them thereafter, assume or state explicidy that God is eternal in the sense that he
exists outside time. On this view, to speak strictly, God does not exist now and did not exist yesterday and will not
exist tomorrow; he just exists in his own timeless moment. By contrast all the biblical writers, almost all theologians
of the first three centuries, some Western theologians of the last few centuries, and (without giving much attention
to the issue) most Eastern Orthodox theologians of all centuries seem to assume that God is in time: he exists at all
moments of time past, present, and future; God is eternal only in the sense that he exists everlastingly. For reasons
given in The Coherence of Theism I prefer the latter view and in general assume it in this book. It makes no difference
to the main argument of the present book which view is adopted, but it would affect the way in which the
argument would need to be expressed at one or two places, if we took the ‘timelessness’ view. For example, the
limit on God's omniscience, that he cannot know what is logically impossible for him to know, would not rule out
his knowledge of the (to us) future actions of free agents. For no actions would be future from God's timeless
perspective. That would have the consequence that the argument in the main body of Chapter 7 against the
possibility of ‘middle knowledge’ would need to be expressed in the way presented in Additional Note 5 below
The Christian understanding of God is more complicated than the general Western understanding (i.e. that
common to Christianity, Judaism, and Islam), in that it conceives of God as ‘three persons’ in one substance, and
that in my view involves these being three persons, each with the cited divine properties. (See my The Christian God
(Clarendon Press, 1994).) But since those persons always act together, that makes no difference to the issues to be
discussed in this book; and so I operate herein with the simpler general Western understanding.

2. [Note 15 to Chapter 4.] The distinction between pleasure and pain, the satisfaction and non-satisfaction of desire,
and a wider conception of good and bad states which include these underlies the distinction between three kinds of
utilitarianism. Hedonistic utilitarianism states

260

ADDITIONAL NOTES

that actions are good in so far as they maximize pleasure and minimize pain. Preference utilitarianism states that
actions are good in so far as they maximize ‘preferences’, which is plausibly interpreted as maximizing the
satisfaction of desires (including, but not confined to, desires to maximize pleasure and minimize pain). Ideal
utilitarianism states that actions are good in so far as they maximize the states which people would choose if they
had perfect information, strong will, settled preferences, etc. But that — according to how it is developed — could
turn out to be no different from the doctrine which states that actions are good in so far as they maximize good
states and minimize bad ones, which is the general form of consequentialism (that actions are good or bad solely in
virtue of their consequences). Now clearly the goodness of its consequences makes for the goodness of an action,
and the badness of its consequences makes for its badness. But I side with the anti-consequentialist in holding that
other considerations are also important in determining the goodness or badness of actions; for example, lying and
some forms of deception are bad, independently of any good consequences they might have. I defend this view at
various places in the book.

3. [Note 3 to Chapter 5.] In thus tying believing an action to be good to having some inclination to do it, I take an
internalist realist view of morality. One alternative to this is moral externalism, the view that believing an action to
be morally good, just like believing an action to have any other property such as causing pain or giving pleasure,
has to be combined with some desire to do an action with that property (and whether a person has such a desire is
a contingent matter) before it inclines us to act. The other alternative is moral internalist anti-realism: the view that
‘believing’ an action to be good is simply being inclined to do it (or approve it, or act in some other way with
respect to it) without holding a belief in any literal sense. Internalist realism about morality seeks to combine the
positive insights of the two approaches to moral philosophy which have characterized the last fifty years. One of
these is the approach of such theories as emotivism and prescriptivism: that holding a moral ‘belief’ is really doing
something or at least being inclined to do something — express an emotion or act in some way and recommend
others to do so. Internalist realism about morality accepts that an inclination to act is involved in moral ‘belief’. But
it also accepts the positive insight of such theories as naturalism and intuitionism that there are moral truths, and
so that ‘beliefs’ about what is morally good or bad really are beliefs which are true or false. To deny that there are
moral truths seems immensely implausible. What can be more obvious than that it really is morally good to feed
the starving and morally bad to torture just for fun? But then I suggest that there would be no

ADDITIONAL NOTES

261

intelligible content to the claim that some action is morally good unless that claim involved the claim that there was
a reason for doing the action. And if someone believes that there is a reason for doing some action, other things
being equal that gives him some inclination to do it.

Among internalist realists about morality are Richard Price, A Review of the Principal Questions of Morals (3rd edn.,
1787), ed. D. D. Raphael (Clarendon Press, 1974); and W. D. Falk, Ought, Reasons, and Morality (Cornell University
Press, 1986), ch. 1. For recent full discussions of rival viewpoints see D. O. Brink, Moral Realism and the Foundations
of Ethics (Cambridge University Press, 1989), ch. 3; and Michael Smith, The Moral Problem (Blackwell, 1994).

4. [Note 13 to Chapter 7.] Plantinga thus affirms (see God, Freedom and Evil (George Allen & Unwin, 1974), 65-73)
(given that God's essential omniscience includes knowledge of the truth-value of all propositions about the future)
that: (1)

God existed at / , and God believed at / that Jones would do X at t^ and it was within Jones's power to refrain
from doing X at ^

entails only (2)

It was within Jones's power [at C] to do something such that, if he had done it, then a belief that God did hold at
/ would have been false

but not (3)

It was within Jones's power at / to do something that would have brought it about that God had a false belief at
U

The difference between (2) and (3) is this. (2) says only that, if Jones did a certain action at / and God's beliefs
remained the same, then one of them would be false. Plantinga accepts that, because he thinks which action Jones
does makes a difference to which beliefs God has (not just to whether the ones he has are true). (3) claims that if
Jones does a certain action, then one of God's beliefs will be false — no further qualifications. But, given my
argument in the text that no action at ^ can make a difference to which beliefs God has at / , then (2) and (3) are
equivalent. So (1) entails (3). God cannot have incorrigibly true beliefs today about what creatures will do freely
tomorrow.

5. [Note 14 to Chapter 7.] The argument in the text against the possibility of God having incorrigible middle
knowledge worked with the assumption that God is in time. It argued that he cannot have essential fore-knowledge
of actual free actions, and so a fortiori of the free actions agents would do in various unrealized circumstances. If we
think of God as outside time, strictly speaking, God would not fore know anything;

262

ADDITIONAL NOTES

for he would not exist before (or simultaneous with, or after) our actions. He would simply exist and know
everything in one timeless moment. In that case Pike's argument cannot be deployed, and I see no reason why God
cannot know the actions which agents will do freely at times (to us) future. Just as we are no less free for the fact
that someone after our actions knows what we have done, so we would be no less free for God seeing in his
timeless moment what we do.

Middle knowledge, however, would still be impossible. For if there is to be a necessary correlation between what
agents do and what God believes about what agents do, it must be sustained either by the actions of agents causing
God's beliefs, or by God causing the actions of agents, or by God's beliefs and the actions of agents having a
common cause. But if the agents have libertarian freedom, nothing causes them to act as they do; and so the last
two alternatives are ruled out. So the correlation can only be sustained by the actions of agents causing God's
beliefs. But while the actual free actions of agents may cause the beliefs of a timeless God and so be compatible
with his essential omniscience, what does not happen can cause nothing. The possible but non-occurrent actions of
actual agents, and the possible but non-occurrent actions of never-to-be-actualized agents can cause nothing. So
God cannot know incorrigibly the truth of counterfactuals of freedom; nor can there ever be a stage at which he
knows incorrigibly what would happen if he were to create a certain sort of world containing free agents. God
cannot have incorrigible middle knowledge, and so ever be in a position to choose to create a world in which, he
knows incorrigibly, free agents will always choose the good.

I do not, however, think that any sense can be made of the supposition that God is outside time. (For argument on
this, see my The Coherence of Theism, rev. edn. (Clarendon Press, 1993), 223-9.) It follows, given Pike's argument, that
God's knowledge at any time of all that it is logically possible to know at that time does not include the future
actions of free agents. On all this, see my The Coherence of Theism, ch. 10, and The Christian God, 131-4. This is
certainly not a traditional view in post-biblical theology, but it does have certain roots in the Old and New
Testaments; see The Coherence of Theism, ch. 10. A very interesting paragraph in a second- (or maybe third-) century
AD theological work which discusses Plantinga's problem is The Clementine Recognitions, 4. 24. The author considers
the objection that God ought ‘not to have created those who, he foresaw, would deviate from the path of
righteousness’. The author's answer is that if God ‘had been afraid of the wickedness of those who were to be’, he
would be ‘like one who could find no other way of remedy or cure, except only this, that he should refrain from his
purpose of creating, lest the wickedness of those

ADDITIONAL NOTES

263

who were to be should be ascribed to him’. This implies that God's foreknowledge is not absolute and that in
creating those likely to ‘deviate from the path of righteousness’, he acts in hope of being able to ‘cure’ them.

6. [Note 3 to Chapter 8.] In recent years John Hick has championed a theodicy which sees the value of suffering in its
providing the opportunity for humans freely to develop their character; the Earth is a ‘vale of soul-making’. See his
Evil and the God ofEove (Macmillan, 1966), pt. 4. He righdy finds a source for this in the writings of Irenaeus. But
Irenaeus seems to think that God could not have made humans perfect from the beginning; humans when first
created were inevitably weak and vulnerable. He writes, somewhat paradoxically, that ‘it was possible for God
himself to have made man perfect from the first, but man could not receive this [perfection], being as yet an infant’
[Adversus Haereses, 4. 38. 1). I cannot see why God could not have made man perfect from the first, but I am
arguing that it is good that humans have the opportunity to choose freely over a significant period whether or not
to become perfect. Hick wants to claim that in the end we shall all through our own free choices become perfect: ‘it
seems morally (although still not logically) impossible that the infinite resourcefulness of infinite love working in
unlimited time should be eternally frustrated, and the creature reject its own good, presented to it in an endless
range of ways’ (p. 380). Maybe. If God always refuses to take no for an answer, we shall yield in the end. Yet if God
is to give someone real freedom to reject him, he must after a finite time take no for an answer. To give us the
choice to reject God, but never to allow that choice to be permanendy executed, is not to give us a real choice at
all.

7. [Note 21 to Chapter 11.] Against this line of argument, John Schellenberg has ( Divine Hiddenness and Human Reason
(Cornell University Press, 1993), ch. 7) a number of brief points. He claims (pp. 198-9) that, given the deep
responsibilities humans already have for future generations, God would not as well give us the further
responsibility for ensuring that they know about God. ‘A loving God’, he writes (p. 199), ‘would not give anyone
the opportunity to put others in a position where neither explicit acceptance nor explicit rejection is possible. God,
if he exists, is concerned to make it possible for each human being, at all times at which she or he exists and is
capable, to be personally related to himself.’ I believe that, with some exceptions, everyone has the opportunity
over life to form their character for good or ill. They may not have reason to believe that there is a God. But if they
have formed their character for good, they will have a natural propensity to show gratitude and respect when it is
due, and so be ready to worship their creator, if they learn of his existence. They may not have that

264

ADDITIONAL NOTES

opportunity until after death, but I see no adequate grounds for supposing that God would ensure that they have it
at every moment of time, in view of the benefits which flow from their not having it for some time. (As for the
‘some exceptions’, clearly the afterlife is needed, either for them to form their character finally or to allow God to
impose upon those who have not formed a character at all a good one.)

If missionary work is a good thing, then, Schellenberg claims (p. 196), missionaries could still have a role in dealing
with culpable non-belief: those who have deceived themselves into believing that there is no God, without there
being any non-culpable believers. So they could. But if missionary work is a good thing, it would be a pity if it
required wrongdoing to set it off. Schellenberg also suggests (p. 179) that the goodness of intellectual inquiry into
divine matters could be satisfied by not making any revelation too clear (not making it too clear where and how
God has acted in history, what he has told us about himself and about ourselves, and how we ought to behave),
while leaving his presence fairly evident. But of course these two are clearly connected: if we already know that
there is a God, then it is much more reasonable to suppose that God would intervene in history, and we require
much less by way of evidence that he has done so. Anyway, it is doubt about God's existence which leads to most
of the temptations to wrongdoing as such having the force they do — by the previous argument. If we know there
is a God, there is not too much temptation not to find out what in detail he has told us in revelation about how we
should act. Failure to do so is apt to indicate doubt about whether he really minds about how we act, i.e. about
whether he is perfectly good, i.e. about whether ‘God’ really is God.

Index

actions, goodness of 82-102; joint and co-operative 94, 164,
211; in response to bad states 96-7, 160-73; seek the good
86, 134—5, 251; see also God, perfect goodness of, and
temptation; spontaneous, defined 84
Acts of the Apostles 101 n., 237, 247 n.

Adam 35-41, 108-10, 116, 119, 210, 216 n, 242
Adams, R. M. 129-31
adults, God's decisions about 233—4
advance directives 226-7, 230, 234

after-life 212, 239, 247-8; as compensation 232, 235-6; see also

Beatific Vision and damnation and Limbo

agnosticism, value of 120, 193, 203-12, 214, 257-8

Alston, W. P. 28, 230 n.

angels 36, 100, 107-8, 116, 242

anger 76

animals xii, 79, 91, 100, 105, 171-5, 189-92, 217-19, 232; see
also fawn caught in fire
apology 97

Aquinas, St Thomas 9 n., 31 n., 35, 38, 80-1, 100 n., 144, 160
n, 162 n., 197 n.

arguments, deductive, defined 7; inductive 7, 59—62, 185—8

Aristotle 65, 92, 244

Athanasius, St 39

Atonement, the, see Christ Jesus

Augustine, St 32, 35, 37 n, 38-41, 43, 52

bad, defined 4—6; incorrigibly, see incorrigibly bad

Barth, K. 32

basing condition for knowledge 57, 63—4, 177

Beatific Vision, the 80, 119-22, 145, 194, 197, 207-9, 232,

250-1

beauty 49-53

being of use 101-5, 167, 231, 235, 241, 244-8

beliefs 15-29, 55-66, 76-80; justification of 15-29, 59-63;

justified , defined 59; justified , defined 59
best of all possible worlds 8
Biel G. 33 n.

Brentano, F. 244
Brink, D. O. 254
Brock, D. 225-8
Brummer, V. 195
Buchanan, A. 225—8
Buddhism xii n.

Butler, J. 211
carers 224-9, 233
causation, backward 132
causing, goodness of 72—3

character formation 91-3, 143-4, 154—5, 167-71, 201, 257,

258

childbirth, thought experiment 242
Chisholm, R. M. 57 n, 238 n.

Christ, Jesus xi, 42, 53, 101, 115 n., 146 n, 213, 215-16, 242,

247, 248, 250

church, the, see revelation

Clementine Recognitions 256-7

Coakley, S. v

Cobb, J. 31

comparative condition 12-13, 237-49

compassion 97, 161-2, 165, 174, 218

competence of dependants 225-6, 233-4

conscience, elimination of 92, 144—5; see also incorrigibly bad

consequentialism 253

Council, Fourth Lateran 36 n.; Jassy 39 n.; Jerusalem 40 n.;
Trent 35,38

counterfactuals of freedom 129—33
courage 162-7, 169, 171-3, 212-13, 217
Credulity, Principle of 20-2, 25-7
damnation 120-2, 143-5, 197-201, 207-9, 250, 257
danger, 172

266

INDEX

Dante 55 n.

Davies, B. 7 n.

Dawkins, M. 174 n.

death 37-8, 40 n, 120, 146, 150, 172, 212-15, 232; see also
Beatific Vision and Hell and life after death and suicide
decay 156-8, 166
deification 119, 251 n.

Dennett, D. 33 n.

dependence on others 93-4, 147-52, 224-33
depravity, transworld 129-33
Descartes, R. 140 n.

desires 67-81, 149-50, 160-2; deterrent and incentive 74,
152-6, 198-203; for life and well-being 75, 153, 207-8, 212,
240—1; modification of, see character formation; to be thought
well of 75, 153, 205—7; see also temptation
determinism 116—18
Deuteronomy, Book of 138
Devil, the 36, 41, 108, 206 n.

Dionysus, Pseudo- 31
Dives and Lazarus 232
divorce 168

Dositheus, confession of 40
Dostversky, F. 246 n.

Draper, P. 20 n.
dying for one's country 102
Ecclesiasticus, Book of 56 n., 125
Eden world 156-8
embodiment 37, 43-4, 48-9
Emerson, F. W. 248

emotions 76-81, 149-50, 160-2 see also love
enjoyment, see pleasure
entailment, defined 19
Ephrem, St 146 n.

epistemic distance from God, need for 250—2

epistemic justification, see justification, epistemic

epistemic use of verbs 20, 21 n., 25—7

evangelism 211, 258

Eve 36, 210

Exodus, Book of 143

expected value, defined 12 n.

Exultet, the 21 6

facts, hard and soft 132—3

failure, utility of 154—5

fairness 236

Falk, W. D. 254

Fall, see Adam and angels

Farren, A. M. 174 n.

fawn caught in fire 103, 171—2, 190, 217, 241, 246
foreknowledge, impossibility of incorrigible 131-2, 255-6
forgiveness 97, 164—5
foundationalism, classical 21 n.

Frankfurt, H. 33 n.

Freddoso, A. F. 128 n.

free will, actuality of human 105-6, 116; defence 11, 33-5,

1 27- 34, 250; different kinds defined 1 1 , 33-4, 84—9; goodness
of 84-9, 241, 248, 250-1; theological understanding of

33-5

friendship 114
Gardiner, R. 118 n.

Genesis, Book of 30, 36, 38, 41, 49-50, 54, 156, 194, 210
Gettier, E. L. 63 n.; effect 63

God defined 3; eternal 252, 255-7; friendship with 112-15;
see also Beatific Vision; inadequacy of language for 113;
omnipotent 3, 8, 30—1, 126—7; omniscient 3—4, 12 n.

128- 34, 231-2, 255; perfectly good 3, 6-7, 9-10; right of 14,
145, 223—36; Trinitarian 253

Goldman A. I. 57 n.

Goodman, N. 60 n.
goodness 4—7, 18, 50-1, 83
Gracia, J. J. N. 32 n.
gratitude 97, 112, 162, 165, 170
greater good defence, defined 1 3
Greene, G. 145 n.

Greer, R. A. 35 n.

Gregory of Nyssa, St 41, 106
Gregory Palamas, St 39 n., 106
grief 162, 212
Griffin, D. R. 31 n.

Guilt, Original 37 n., 38 n.
hardhearted, the 169-70, 199-200, 245

INDEX

267

Hartshorne, C. 31
Hasker, W. 131
Heaven, see Beatific Vision
Hegel, G. W. F. 32
Heidegger, M. 32
Hell, see damnation
Hick, J. 256
Hickey, D. 118 n.

higher-order goods defence 162—75
Hinduism xii n.

Holocaust, the 107, 151-2,

Honesty, Principle of 138-40, 155, 165, 177, 181, 200, 218
Horace 102 n.

Howard-Snyder, D. 28 n.
image of God 65, 106

incompetence, see competence of dependants

incorrigibly bad, the 92, 120-1, 143-4, 197, 199-201, 208, 245

indeterminism 116—18

inquiry, religious, goodness of 194, 210—12

Irenaeus, St 39, 109, 256

Isaiah, the Solitary, St 210 n.

James, St, Epistle of 145
Jeremiah 200 n.

Job, Book of 3, 36, 41

John, Gospel of St 42, 160, 213, 247 n.

John of Damascus, St 65, 106, 170 n.

John, Revelation of St 111, 251
justice 142

justification, epistemic 15-27, 59-66, 180-92

justified , defined 59

Kant, I. 2 11 3, 233

killing 146, 150, 172, 212, 214

Kirk, K. E. 80 n.

knowledge 57-66, 75, 90-1, 141-3, 177-92; defence from the
need for 176-92; middle 128-33,254-6
Korcz, K. A. 63 n.

laws of nature 51, 98-9, 116-18, 128, 185-8, 190-1
learning from experience 65—6, 153, 182—92, 217; see also
knowledge

Lecky, W E. H. 102 n.

Lehrer, K. 58 n., 63 n.

Leibniz, G. W. 8 n., 9 n., 32
Lemos, N. 244
Lessing, G E. 98 n.

Leviticus, Book of 143
Lewis, C. S. 202 n.

Lewis, D. 130

life after death, see after-life goodness of 240-1

Limbo 121, 232

logical impossibility 125-7

love 77-9, 100-1, 114, 167, 234; of God 114, 194-5, 205-7;

see also Beatific Vision
Lovejoy, A. O. 100 n.

Luke, Gospel of St 232 n., 248 n., 250 n.

Luther, M. 33 n.

MacDonald, S. 52 n.

Manichaeism 30
Mark, Gospel of St 247 n.

Matthew, Gospel of St 42 n., 53 n, 79 n, 115 n., 210, 247, 248

Maximus the Confessor, St 251 n.
memory, goodness of 55—6
mental, defined 54 n.
mercy 142

Mill, J. S. 70 n., 242, 244
miracles 117-18
Molina, L. de 128 n.

Moltmann, J. 42 n.

Monte Carlo fallacy 59
Moore, G E. 18 n.

moral belief, importance and nature of 5—6, 18—19, 57, 85—6,

91-2, 141-5, 168-9, 178-9, 253-4

moral evil, defined 4

Moser, P. K 178 n.

natural affection, see love

natural evil, defined 5

Neely, W. 33 n.

nichtige, das 32

obedience 228

obligation 5, 10, 73; see also sin and sinfulness
O'Connor, D. 1 77 n.

Ogden, S. 31

oppression of one group by another 166
Origen xii

268

INDEX

pain 68, 74, 163-4,

Parfit, D. 244

patience 162-4, 170-1, 213

Paul St:; Epistle to Colossians 247 n.; Epistle to Corinthians,
First 176, 209, 247 n.; Second 115 n.; Epistle to Romans
36-7, 42, 165 n., 209, 211 n, 215 n., 223, 237
Pelagius 40
Peter of Blois 202-3,

Peter of Damascus, St 146 n., 170 £, 198 n.

Peter of Moghila 39 n.

Philip the Chancellor 144

Philokalia 1 70, 233; see also Gregory Palamas, St and Peter of
Damascus, St
Pike, N. 131, 255

Plantinga, A. 15 n, 16 n, 21 n, 58 n, 127-33, 254, 255

Plato 65, 100, 102 n., 240

pleasure 67—73

plenitude, principle of 100

poverty 170—1

prayer 115, 118, 203-4

Price, R. 254

pride 104, 168, 202

privatio boni 8, 31—2

probability, epistemic 59—62, 172 n.; see also justification,
epistemic; natural or physical 116, 172 n.
problem of evil, defined 3-14, 20 n.

Process Theology 31
projectible predicates 60 n.
promises 94-7, 244
proper functioning 57 n., 58

providence, general and special ix, 116 mechanism of 116—18
punishment 142, 152-3, 155-6, 196; after this life 121-2,
197-201, 208; see also damnation; for sin, in this life 193,
196-7, 236
Purgatory 197, 201
Quantum Theory 116

rational and rational , see justified and justified
reliabilism 57—8
reparation 97, 164—5
respect 76, 165

responsibility, scope of 89-101, 141-54, 159
result of action, defined 181

revelation x-xii, 106-7, 115, 118, 120, 122, 219, 246
Revelation of St John, see John, Revelation of St
reward 79 n., 97 n., 165, 247-8

right, defined 5, 223; of carers 224—9; of God to allow bad

10-11, 14, 213-14, 223-36, 238, 242, 249

Ross, W. D. 244

Rowe, W. L. 26 n., 103 n.

Satan, see Devil, the
Schellenberg J. L. 204 n., 256-7
S chleiermacher, F. 43

Schlesinger, G 8 n.

Scotus, Duns 33 n., 35, 38
self-deception 55, 203-4

self-determination, good to be weighed 224—35; see also free

will, goodness of and responsibility

self-pity 162, 217

self-sacrifice 212-14

Sermon on the Mount 143, 237

service to others, see being of use

sexual abuse 228—9

Shaw, G. B. 200

sin 112, 193—5; mortal and venial 136, 209; Original 36—41;
see also Adam

sinfulness 36-41 ; see also temptation
slave trade 254—6
sloth 152, 156-8
Smith, D. 118 n.

Smith, M. 254
Socrates 240

solidarity in suffering 216

soul formation, see character formation

Stump, E. v, 177 n., 202, 230 n.

Suarez, F. 32 n.

substituted judgement 226, 234
suffering, defined 69

suicide 146-7, 150, 153-4, 212, 234-5, 240-1

Swinburne, R. G. v, x, xi n., 6 n., 1 5 n., 20 n., 24 n., 59 n., 61 n.,

118 n., 132 n., 146 n., 172 n, 177 n, 208 n, 255

INDEX

269

"Evolution of the Soul 54 n., 67 n., 90 n., 106 n., 116 n.

"Responsibility and Atonement 10 n., 35 n., 37 n., 40 n., 83 n., 84 n., 95 n., 97 n., 112 n., 119 n., 142 n., 144 n., 145 n., 146 n., 194 n.,
196 n., 201 n., 215 n.

Symeon, St, The New Theologian 199 n.

sympathy 96-7, 162-9, 171

tares and wheat, parable 42 n.

temptation 86, 108, 134—7, 145-6, 156-8, 169

Tennant, F. R. 43-4

testimony 61, 184, 188

theodicy x-xi, 14-29, 44-5, 204 n, 237, 249-50
thought experiments 240-6
toy world 242-3

unborn, God's decisions about 229-33
unemployment 102

unselfish decisions for others 227, 231
use, being of, see being of use
utilitarianism 252—3
van Inwagen, P. 28 n.

Venantius Fortunatus 248
vocation 115, 194
vows 95-6, 155-6, 199
warrant 57
Watson, G. 33 n.
wealth 170-1

weighing good against bad, see comparative condition
well-being 200, 208, 227, 234
Whitehead, A. N. 31
Williams, N. P. 39 n., 40 n.

Williams, R. 119 n.

Wisdom of Solomon 56 n., 104 n., 202 n.

Wolterstorff, N. 32 n.
world, end of 249

worship 112, 202; see also Beatific Vision
wrong, defined 5, 10 n.

Wykstra, S. 25-9

young inhibited by old 213—14

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "RichardSwinburne"

See other formats